diff options
author | Renato Botelho <renato@netgate.com> | 2016-08-17 15:23:38 -0300 |
---|---|---|
committer | Renato Botelho <renato@netgate.com> | 2016-08-17 15:23:38 -0300 |
commit | 75cd8d40056c799f03b759475d9bfd10ba266a6c (patch) | |
tree | 60433235501684bffeab90e65139a8285fcf46a9 | |
parent | 99990a0d149f0eae805aa1f49d4a61be30c3b000 (diff) | |
parent | ad413762f28e3be343987e707b9cf4f10f963693 (diff) | |
download | FreeBSD-src-75cd8d40056c799f03b759475d9bfd10ba266a6c.zip FreeBSD-src-75cd8d40056c799f03b759475d9bfd10ba266a6c.tar.gz |
Merge remote-tracking branch 'origin/stable/10' into devel
51 files changed, 1716 insertions, 559 deletions
diff --git a/bin/ps/ps.1 b/bin/ps/ps.1 index 17bfa41..fdb4f1e 100644 --- a/bin/ps/ps.1 +++ b/bin/ps/ps.1 @@ -29,7 +29,7 @@ .\" @(#)ps.1 8.3 (Berkeley) 4/18/94 .\" $FreeBSD$ .\" -.Dd December 9, 2014 +.Dd August 12, 2016 .Dt PS 1 .Os .Sh NAME @@ -310,7 +310,6 @@ the include file .It Dv "P_ADVLOCK" Ta No "0x00001" Ta "Process may hold a POSIX advisory lock" .It Dv "P_CONTROLT" Ta No "0x00002" Ta "Has a controlling terminal" .It Dv "P_KTHREAD" Ta No "0x00004" Ta "Kernel thread" -.It Dv "P_FOLLOWFORK" Ta No "0x00008" Ta "Attach debugger to new children" .It Dv "P_PPWAIT" Ta No "0x00010" Ta "Parent is waiting for child to exec/exit" .It Dv "P_PROFIL" Ta No "0x00020" Ta "Has started profiling" .It Dv "P_STOPPROF" Ta No "0x00040" Ta "Has thread in requesting to stop prof" @@ -348,6 +347,7 @@ the include file .In sys/proc.h : .Bl -column P2_INHERIT_PROTECTED 0x00000001 .It Dv "P2_INHERIT_PROTECTED" Ta No "0x00000001" Ta "New children get P_PROTECTED" +.It Dv "P2_PTRACE_FSTP" Ta No "0x00000010" Ta "SIGSTOP from PT_ATTACH not yet handled" .El .It Cm label The MAC label of the process. diff --git a/cddl/contrib/opensolaris/cmd/zdb/zdb.8 b/cddl/contrib/opensolaris/cmd/zdb/zdb.8 index 4ef437e..6ae25c3 100644 --- a/cddl/contrib/opensolaris/cmd/zdb/zdb.8 +++ b/cddl/contrib/opensolaris/cmd/zdb/zdb.8 @@ -86,6 +86,17 @@ pool, and is inherently unstable. The precise output of most invocations is not documented, a knowledge of ZFS internals is assumed. .Pp +If the +.Ar dataset +argument does not contain any +.Sy / +or +.Sy @ +characters, it is interpreted as a pool name. +The root dataset can be specified as +.Pa pool Ns Sy / +(pool name followed by a slash). +.Pp When operating on an imported and active pool it is possible, though unlikely, that zdb may interpret inconsistent pool data and behave erratically. .Sh OPTIONS diff --git a/cddl/contrib/opensolaris/cmd/zdb/zdb.c b/cddl/contrib/opensolaris/cmd/zdb/zdb.c index 7c08c5f..c97cd36 100644 --- a/cddl/contrib/opensolaris/cmd/zdb/zdb.c +++ b/cddl/contrib/opensolaris/cmd/zdb/zdb.c @@ -3559,12 +3559,23 @@ main(int argc, char **argv) nvlist_t *policy = NULL; uint64_t max_txg = UINT64_MAX; int rewind = ZPOOL_NEVER_REWIND; + char *spa_config_path_env; + boolean_t target_is_spa = B_TRUE; (void) setrlimit(RLIMIT_NOFILE, &rl); (void) enable_extended_FILE_stdio(-1, -1); dprintf_setup(&argc, argv); + /* + * If there is an environment variable SPA_CONFIG_PATH it overrides + * default spa_config_path setting. If -U flag is specified it will + * override this environment variable settings once again. + */ + spa_config_path_env = getenv("SPA_CONFIG_PATH"); + if (spa_config_path_env != NULL) + spa_config_path = spa_config_path_env; + while ((c = getopt(argc, argv, "bcdhilmMI:suCDRSAFLXx:evp:t:U:P")) != -1) { switch (c) { @@ -3729,8 +3740,23 @@ main(int argc, char **argv) } } + if (strpbrk(target, "/@") != NULL) { + size_t targetlen; + + target_is_spa = B_FALSE; + /* + * Remove any trailing slash. Later code would get confused + * by it, but we want to allow it so that "pool/" can + * indicate that we want to dump the topmost filesystem, + * rather than the whole pool. + */ + targetlen = strlen(target); + if (targetlen != 0 && target[targetlen - 1] == '/') + target[targetlen - 1] = '\0'; + } + if (error == 0) { - if (strpbrk(target, "/@") == NULL || dump_opt['R']) { + if (target_is_spa || dump_opt['R']) { error = spa_open_rewind(target, &spa, FTAG, policy, NULL); if (error) { diff --git a/lib/libc/gen/glob.3 b/lib/libc/gen/glob.3 index 468a73b..398c14b 100644 --- a/lib/libc/gen/glob.3 +++ b/lib/libc/gen/glob.3 @@ -275,24 +275,10 @@ is .Pf non- Dv NULL , .Fn glob calls -.Fa \*(lp*errfunc\*(rp Ns ( Fa path , errno ) . -This may be unintuitive: a pattern like -.Ql */Makefile -will try to -.Xr stat 2 -.Ql foo/Makefile -even if -.Ql foo -is not a directory, resulting in a -call to -.Fa errfunc . -The error routine can suppress this action by testing for -.Er ENOENT -and -.Er ENOTDIR ; +.Fa \*(lp*errfunc\*(rp Ns ( Fa path , errno ) , however, the .Dv GLOB_ERR -flag will still cause an immediate +flag will cause an immediate return when this happens. .Pp If @@ -377,7 +363,7 @@ file .It Dv GLOB_NOSPACE An attempt to allocate memory failed, or if .Fa errno -was 0 +was E2BIG, .Dv GLOB_LIMIT was specified in the flags and .Fa pglob\->gl_matchc diff --git a/lib/libc/gen/glob.c b/lib/libc/gen/glob.c index 6c86297..5df0b74 100644 --- a/lib/libc/gen/glob.c +++ b/lib/libc/gen/glob.c @@ -71,7 +71,7 @@ __FBSDID("$FreeBSD$"); * 1. Patterns with illegal byte sequences match nothing - even if * GLOB_NOCHECK is specified. * 2. Illegal byte sequences in filenames are handled by treating them as - * single-byte characters with a value of the first byte of the sequence + * single-byte characters with a values of such bytes of the sequence * cast to wchar_t. * 3. State-dependent encodings are not currently supported. */ @@ -113,25 +113,20 @@ struct glob_limit { size_t l_string_cnt; }; -#define DOLLAR '$' -#define DOT '.' -#define EOS '\0' -#define LBRACKET '[' -#define NOT '!' -#define QUESTION '?' -#define QUOTE '\\' -#define RANGE '-' -#define RBRACKET ']' -#define SEP '/' -#define STAR '*' -#define TILDE '~' -#define UNDERSCORE '_' -#define LBRACE '{' -#define RBRACE '}' -#define SLASH '/' -#define COMMA ',' - -#ifndef DEBUG +#define DOT L'.' +#define EOS L'\0' +#define LBRACKET L'[' +#define NOT L'!' +#define QUESTION L'?' +#define QUOTE L'\\' +#define RANGE L'-' +#define RBRACKET L']' +#define SEP L'/' +#define STAR L'*' +#define TILDE L'~' +#define LBRACE L'{' +#define RBRACE L'}' +#define COMMA L',' #define M_QUOTE 0x8000000000ULL #define M_PROTECT 0x4000000000ULL @@ -140,28 +135,19 @@ struct glob_limit { typedef uint_fast64_t Char; -#else - -#define M_QUOTE 0x80 -#define M_PROTECT 0x40 -#define M_MASK 0xff -#define M_CHAR 0x7f - -typedef char Char; - -#endif - - #define CHAR(c) ((Char)((c)&M_CHAR)) #define META(c) ((Char)((c)|M_QUOTE)) -#define M_ALL META('*') -#define M_END META(']') -#define M_NOT META('!') -#define M_ONE META('?') -#define M_RNG META('-') -#define M_SET META('[') +#define UNPROT(c) ((c) & ~M_PROTECT) +#define M_ALL META(L'*') +#define M_END META(L']') +#define M_NOT META(L'!') +#define M_ONE META(L'?') +#define M_RNG META(L'-') +#define M_SET META(L'[') #define ismeta(c) (((c)&M_QUOTE) != 0) - +#ifdef DEBUG +#define isprot(c) (((c)&M_PROTECT) != 0) +#endif static int compare(const void *, const void *); static int g_Ctoc(const Char *, char *, size_t); @@ -172,19 +158,27 @@ static const Char *g_strchr(const Char *, wchar_t); static Char *g_strcat(Char *, const Char *); #endif static int g_stat(Char *, struct stat *, glob_t *); -static int glob0(const Char *, glob_t *, struct glob_limit *); +static int glob0(const Char *, glob_t *, struct glob_limit *, + const char *); static int glob1(Char *, glob_t *, struct glob_limit *); static int glob2(Char *, Char *, Char *, Char *, glob_t *, struct glob_limit *); static int glob3(Char *, Char *, Char *, Char *, Char *, glob_t *, struct glob_limit *); -static int globextend(const Char *, glob_t *, struct glob_limit *); +static int globextend(const Char *, glob_t *, struct glob_limit *, + const char *); static const Char * globtilde(const Char *, Char *, size_t, glob_t *); +static int globexp0(const Char *, glob_t *, struct glob_limit *, + const char *); static int globexp1(const Char *, glob_t *, struct glob_limit *); -static int globexp2(const Char *, const Char *, glob_t *, int *, +static int globexp2(const Char *, const Char *, glob_t *, struct glob_limit *); +static int globfinal(glob_t *, struct glob_limit *, size_t, + const char *); static int match(Char *, Char *, Char *); +static int err_nomatch(glob_t *, struct glob_limit *, const char *); +static int err_aborted(glob_t *, int, char *); #ifdef DEBUG static void qprintf(const char *, Char *); #endif @@ -199,6 +193,7 @@ glob(const char * __restrict pattern, int flags, mbstate_t mbs; wchar_t wc; size_t clen; + int too_long; patnext = pattern; if (!(flags & GLOB_APPEND)) { @@ -218,24 +213,27 @@ glob(const char * __restrict pattern, int flags, bufnext = patbuf; bufend = bufnext + MAXPATHLEN - 1; + too_long = 1; if (flags & GLOB_NOESCAPE) { memset(&mbs, 0, sizeof(mbs)); - while (bufend - bufnext >= MB_CUR_MAX) { + while (bufnext <= bufend) { clen = mbrtowc(&wc, patnext, MB_LEN_MAX, &mbs); if (clen == (size_t)-1 || clen == (size_t)-2) - return (GLOB_NOMATCH); - else if (clen == 0) + return (err_nomatch(pglob, &limit, pattern)); + else if (clen == 0) { + too_long = 0; break; + } *bufnext++ = wc; patnext += clen; } } else { /* Protect the quoted characters. */ memset(&mbs, 0, sizeof(mbs)); - while (bufend - bufnext >= MB_CUR_MAX) { - if (*patnext == QUOTE) { - if (*++patnext == EOS) { - *bufnext++ = QUOTE | M_PROTECT; + while (bufnext <= bufend) { + if (*patnext == '\\') { + if (*++patnext == '\0') { + *bufnext++ = QUOTE; continue; } prot = M_PROTECT; @@ -243,19 +241,47 @@ glob(const char * __restrict pattern, int flags, prot = 0; clen = mbrtowc(&wc, patnext, MB_LEN_MAX, &mbs); if (clen == (size_t)-1 || clen == (size_t)-2) - return (GLOB_NOMATCH); - else if (clen == 0) + return (err_nomatch(pglob, &limit, pattern)); + else if (clen == 0) { + too_long = 0; break; + } *bufnext++ = wc | prot; patnext += clen; } } + if (too_long) + return (err_nomatch(pglob, &limit, pattern)); *bufnext = EOS; if (flags & GLOB_BRACE) - return (globexp1(patbuf, pglob, &limit)); + return (globexp0(patbuf, pglob, &limit, pattern)); else - return (glob0(patbuf, pglob, &limit)); + return (glob0(patbuf, pglob, &limit, pattern)); +} + +static int +globexp0(const Char *pattern, glob_t *pglob, struct glob_limit *limit, + const char *origpat) { + int rv; + size_t oldpathc; + + /* Protect a single {}, for find(1), like csh */ + if (pattern[0] == LBRACE && pattern[1] == RBRACE && pattern[2] == EOS) { + if ((pglob->gl_flags & GLOB_LIMIT) && + limit->l_brace_cnt++ >= GLOB_LIMIT_BRACE) { + errno = E2BIG; + return (GLOB_NOSPACE); + } + return (glob0(pattern, pglob, limit, origpat)); + } + + oldpathc = pglob->gl_pathc; + + if ((rv = globexp1(pattern, pglob, limit)) != 0) + return rv; + + return (globfinal(pglob, limit, oldpathc, origpat)); } /* @@ -266,24 +292,18 @@ glob(const char * __restrict pattern, int flags, static int globexp1(const Char *pattern, glob_t *pglob, struct glob_limit *limit) { - const Char* ptr = pattern; - int rv; + const Char* ptr; - if ((pglob->gl_flags & GLOB_LIMIT) && - limit->l_brace_cnt++ >= GLOB_LIMIT_BRACE) { - errno = 0; - return (GLOB_NOSPACE); + if ((ptr = g_strchr(pattern, LBRACE)) != NULL) { + if ((pglob->gl_flags & GLOB_LIMIT) && + limit->l_brace_cnt++ >= GLOB_LIMIT_BRACE) { + errno = E2BIG; + return (GLOB_NOSPACE); + } + return (globexp2(ptr, pattern, pglob, limit)); } - /* Protect a single {}, for find(1), like csh */ - if (pattern[0] == LBRACE && pattern[1] == RBRACE && pattern[2] == EOS) - return glob0(pattern, pglob, limit); - - while ((ptr = g_strchr(ptr, LBRACE)) != NULL) - if (!globexp2(ptr, pattern, pglob, &rv, limit)) - return rv; - - return glob0(pattern, pglob, limit); + return (glob0(pattern, pglob, limit, NULL)); } @@ -293,10 +313,10 @@ globexp1(const Char *pattern, glob_t *pglob, struct glob_limit *limit) * If it fails then it tries to glob the rest of the pattern and returns. */ static int -globexp2(const Char *ptr, const Char *pattern, glob_t *pglob, int *rv, +globexp2(const Char *ptr, const Char *pattern, glob_t *pglob, struct glob_limit *limit) { - int i; + int i, rv; Char *lm, *ls; const Char *pe, *pm, *pm1, *pl; Char patbuf[MAXPATHLEN]; @@ -308,7 +328,7 @@ globexp2(const Char *ptr, const Char *pattern, glob_t *pglob, int *rv, ls = lm; /* Find the balanced brace */ - for (i = 0, pe = ++ptr; *pe; pe++) + for (i = 0, pe = ++ptr; *pe != EOS; pe++) if (*pe == LBRACKET) { /* Ignore everything between [] */ for (pm = pe++; *pe != RBRACKET && *pe != EOS; pe++) @@ -330,10 +350,8 @@ globexp2(const Char *ptr, const Char *pattern, glob_t *pglob, int *rv, } /* Non matching braces; just glob the pattern */ - if (i != 0 || *pe == EOS) { - *rv = glob0(patbuf, pglob, limit); - return (0); - } + if (i != 0 || *pe == EOS) + return (glob0(pattern, pglob, limit, NULL)); for (i = 0, pl = pm = ptr; pm <= pe; pm++) switch (*pm) { @@ -378,7 +396,9 @@ globexp2(const Char *ptr, const Char *pattern, glob_t *pglob, int *rv, #ifdef DEBUG qprintf("globexp2:", patbuf); #endif - *rv = globexp1(patbuf, pglob, limit); + rv = globexp1(patbuf, pglob, limit); + if (rv) + return (rv); /* move after the comma, to the next string */ pl = pm + 1; @@ -388,7 +408,6 @@ globexp2(const Char *ptr, const Char *pattern, glob_t *pglob, int *rv, default: break; } - *rv = 0; return (0); } @@ -401,9 +420,15 @@ static const Char * globtilde(const Char *pattern, Char *patbuf, size_t patbuf_len, glob_t *pglob) { struct passwd *pwd; - char *h; + char *h, *sc; const Char *p; Char *b, *eb; + wchar_t wc; + wchar_t wbuf[MAXPATHLEN]; + wchar_t *wbufend, *dc; + size_t clen; + mbstate_t mbs; + int too_long; if (*pattern != TILDE || !(pglob->gl_flags & GLOB_TILDE)) return (pattern); @@ -412,13 +437,17 @@ globtilde(const Char *pattern, Char *patbuf, size_t patbuf_len, glob_t *pglob) * Copy up to the end of the string or / */ eb = &patbuf[patbuf_len - 1]; - for (p = pattern + 1, h = (char *) patbuf; - h < (char *)eb && *p && *p != SLASH; *h++ = *p++) + for (p = pattern + 1, b = patbuf; + b < eb && *p != EOS && UNPROT(*p) != SEP; *b++ = *p++) continue; - *h = EOS; + if (*p != EOS && UNPROT(*p) != SEP) + return (NULL); + + *b = EOS; + h = NULL; - if (((char *) patbuf)[0] == EOS) { + if (patbuf[0] == EOS) { /* * handle a plain ~ or ~/ by expanding $HOME first (iff * we're not running setuid or setgid) and then trying @@ -438,20 +467,56 @@ globtilde(const Char *pattern, Char *patbuf, size_t patbuf_len, glob_t *pglob) /* * Expand a ~user */ - if ((pwd = getpwnam((char*) patbuf)) == NULL) + if (g_Ctoc(patbuf, (char *)wbuf, sizeof(wbuf))) + return (NULL); + if ((pwd = getpwnam((char *)wbuf)) == NULL) return (pattern); else h = pwd->pw_dir; } /* Copy the home directory */ - for (b = patbuf; b < eb && *h; *b++ = *h++) + dc = wbuf; + sc = h; + wbufend = wbuf + MAXPATHLEN - 1; + too_long = 1; + memset(&mbs, 0, sizeof(mbs)); + while (dc <= wbufend) { + clen = mbrtowc(&wc, sc, MB_LEN_MAX, &mbs); + if (clen == (size_t)-1 || clen == (size_t)-2) { + /* XXX See initial comment #2. */ + wc = (unsigned char)*sc; + clen = 1; + memset(&mbs, 0, sizeof(mbs)); + } + if ((*dc++ = wc) == EOS) { + too_long = 0; + break; + } + sc += clen; + } + if (too_long) + return (NULL); + + dc = wbuf; + for (b = patbuf; b < eb && *dc != EOS; *b++ = *dc++ | M_PROTECT) continue; + if (*dc != EOS) + return (NULL); /* Append the rest of the pattern */ - while (b < eb && (*b++ = *p++) != EOS) - continue; - *b = EOS; + if (*p != EOS) { + too_long = 1; + while (b <= eb) { + if ((*b++ = *p++) == EOS) { + too_long = 0; + break; + } + } + if (too_long) + return (NULL); + } else + *b = EOS; return (patbuf); } @@ -464,14 +529,18 @@ globtilde(const Char *pattern, Char *patbuf, size_t patbuf_len, glob_t *pglob) * if things went well, nonzero if errors occurred. */ static int -glob0(const Char *pattern, glob_t *pglob, struct glob_limit *limit) -{ +glob0(const Char *pattern, glob_t *pglob, struct glob_limit *limit, + const char *origpat) { const Char *qpatnext; int err; size_t oldpathc; Char *bufnext, c, patbuf[MAXPATHLEN]; qpatnext = globtilde(pattern, patbuf, MAXPATHLEN, pglob); + if (qpatnext == NULL) { + errno = E2BIG; + return (GLOB_NOSPACE); + } oldpathc = pglob->gl_pathc; bufnext = patbuf; @@ -530,30 +599,29 @@ glob0(const Char *pattern, glob_t *pglob, struct glob_limit *limit) if ((err = glob1(patbuf, pglob, limit)) != 0) return(err); - /* - * If there was no match we are going to append the pattern - * if GLOB_NOCHECK was specified or if GLOB_NOMAGIC was specified - * and the pattern did not contain any magic characters - * GLOB_NOMAGIC is there just for compatibility with csh. - */ - if (pglob->gl_pathc == oldpathc) { - if (((pglob->gl_flags & GLOB_NOCHECK) || - ((pglob->gl_flags & GLOB_NOMAGIC) && - !(pglob->gl_flags & GLOB_MAGCHAR)))) - return (globextend(pattern, pglob, limit)); - else - return (GLOB_NOMATCH); - } + if (origpat != NULL) + return (globfinal(pglob, limit, oldpathc, origpat)); + + return (0); +} + +static int +globfinal(glob_t *pglob, struct glob_limit *limit, size_t oldpathc, + const char *origpat) { + if (pglob->gl_pathc == oldpathc) + return (err_nomatch(pglob, limit, origpat)); + if (!(pglob->gl_flags & GLOB_NOSORT)) qsort(pglob->gl_pathv + pglob->gl_offs + oldpathc, pglob->gl_pathc - oldpathc, sizeof(char *), compare); + return (0); } static int compare(const void *p, const void *q) { - return (strcmp(*(char **)p, *(char **)q)); + return (strcoll(*(char **)p, *(char **)q)); } static int @@ -593,44 +661,47 @@ glob2(Char *pathbuf, Char *pathend, Char *pathend_last, Char *pattern, if ((pglob->gl_flags & GLOB_LIMIT) && limit->l_stat_cnt++ >= GLOB_LIMIT_STAT) { - errno = 0; - if (pathend + 1 > pathend_last) - return (GLOB_ABORTED); - *pathend++ = SEP; - *pathend = EOS; + errno = E2BIG; return (GLOB_NOSPACE); } - if (((pglob->gl_flags & GLOB_MARK) && - pathend[-1] != SEP) && (S_ISDIR(sb.st_mode) - || (S_ISLNK(sb.st_mode) && - (g_stat(pathbuf, &sb, pglob) == 0) && + if ((pglob->gl_flags & GLOB_MARK) && + UNPROT(pathend[-1]) != SEP && + (S_ISDIR(sb.st_mode) || + (S_ISLNK(sb.st_mode) && + g_stat(pathbuf, &sb, pglob) == 0 && S_ISDIR(sb.st_mode)))) { - if (pathend + 1 > pathend_last) - return (GLOB_ABORTED); + if (pathend + 1 > pathend_last) { + errno = E2BIG; + return (GLOB_NOSPACE); + } *pathend++ = SEP; *pathend = EOS; } ++pglob->gl_matchc; - return (globextend(pathbuf, pglob, limit)); + return (globextend(pathbuf, pglob, limit, NULL)); } /* Find end of next segment, copy tentatively to pathend. */ q = pathend; p = pattern; - while (*p != EOS && *p != SEP) { + while (*p != EOS && UNPROT(*p) != SEP) { if (ismeta(*p)) anymeta = 1; - if (q + 1 > pathend_last) - return (GLOB_ABORTED); + if (q + 1 > pathend_last) { + errno = E2BIG; + return (GLOB_NOSPACE); + } *q++ = *p++; } if (!anymeta) { /* No expansion, do next segment. */ pathend = q; pattern = p; - while (*pattern == SEP) { - if (pathend + 1 > pathend_last) - return (GLOB_ABORTED); + while (UNPROT(*pattern) == SEP) { + if (pathend + 1 > pathend_last) { + errno = E2BIG; + return (GLOB_NOSPACE); + } *pathend++ = *pattern++; } } else /* Need expansion, recurse. */ @@ -647,42 +718,44 @@ glob3(Char *pathbuf, Char *pathend, Char *pathend_last, { struct dirent *dp; DIR *dirp; - int err; - char buf[MAXPATHLEN]; + int err, too_long, saverrno, saverrno2; + char buf[MAXPATHLEN + MB_LEN_MAX - 1]; - /* - * The readdirfunc declaration can't be prototyped, because it is - * assigned, below, to two functions which are prototyped in glob.h - * and dirent.h as taking pointers to differently typed opaque - * structures. - */ - struct dirent *(*readdirfunc)(); + struct dirent *(*readdirfunc)(DIR *); - if (pathend > pathend_last) - return (GLOB_ABORTED); + if (pathend > pathend_last) { + errno = E2BIG; + return (GLOB_NOSPACE); + } *pathend = EOS; - errno = 0; + if (pglob->gl_errfunc != NULL && + g_Ctoc(pathbuf, buf, sizeof(buf))) { + errno = E2BIG; + return (GLOB_NOSPACE); + } + saverrno = errno; + errno = 0; if ((dirp = g_opendir(pathbuf, pglob)) == NULL) { - /* TODO: don't call for ENOENT or ENOTDIR? */ - if (pglob->gl_errfunc) { - if (g_Ctoc(pathbuf, buf, sizeof(buf))) - return (GLOB_ABORTED); - if (pglob->gl_errfunc(buf, errno) || - pglob->gl_flags & GLOB_ERR) - return (GLOB_ABORTED); - } - return (0); + if (errno == ENOENT || errno == ENOTDIR) + return (0); + err = err_aborted(pglob, errno, buf); + if (errno == 0) + errno = saverrno; + return (err); } err = 0; - /* Search directory for matching names. */ + /* pglob->gl_readdir takes a void *, fix this manually */ if (pglob->gl_flags & GLOB_ALTDIRFUNC) - readdirfunc = pglob->gl_readdir; + readdirfunc = (struct dirent *(*)(DIR *))pglob->gl_readdir; else readdirfunc = readdir; - while ((dp = (*readdirfunc)(dirp))) { + + errno = 0; + /* Search directory for matching names. */ + while ((dp = (*readdirfunc)(dirp)) != NULL) { char *sc; Char *dc; wchar_t wc; @@ -691,49 +764,70 @@ glob3(Char *pathbuf, Char *pathend, Char *pathend_last, if ((pglob->gl_flags & GLOB_LIMIT) && limit->l_readdir_cnt++ >= GLOB_LIMIT_READDIR) { - errno = 0; - if (pathend + 1 > pathend_last) - err = GLOB_ABORTED; - else { - *pathend++ = SEP; - *pathend = EOS; - err = GLOB_NOSPACE; - } + errno = E2BIG; + err = GLOB_NOSPACE; break; } /* Initial DOT must be matched literally. */ - if (dp->d_name[0] == DOT && *pattern != DOT) + if (dp->d_name[0] == '.' && UNPROT(*pattern) != DOT) { + errno = 0; continue; + } memset(&mbs, 0, sizeof(mbs)); dc = pathend; sc = dp->d_name; - while (dc < pathend_last) { + too_long = 1; + while (dc <= pathend_last) { clen = mbrtowc(&wc, sc, MB_LEN_MAX, &mbs); if (clen == (size_t)-1 || clen == (size_t)-2) { - wc = *sc; + /* XXX See initial comment #2. */ + wc = (unsigned char)*sc; clen = 1; memset(&mbs, 0, sizeof(mbs)); } - if ((*dc++ = wc) == EOS) + if ((*dc++ = wc) == EOS) { + too_long = 0; break; + } sc += clen; } - if (!match(pathend, pattern, restpattern)) { + if (too_long && (err = err_aborted(pglob, ENAMETOOLONG, + buf))) { + errno = ENAMETOOLONG; + break; + } + if (too_long || !match(pathend, pattern, restpattern)) { *pathend = EOS; + errno = 0; continue; } + if (errno == 0) + errno = saverrno; err = glob2(pathbuf, --dc, pathend_last, restpattern, pglob, limit); if (err) break; + errno = 0; } + saverrno2 = errno; if (pglob->gl_flags & GLOB_ALTDIRFUNC) (*pglob->gl_closedir)(dirp); else closedir(dirp); - return (err); + errno = saverrno2; + + if (err) + return (err); + + if (dp == NULL && errno != 0 && + (err = err_aborted(pglob, errno, buf))) + return (err); + + if (errno == 0) + errno = saverrno; + return (0); } @@ -752,7 +846,8 @@ glob3(Char *pathbuf, Char *pathend, Char *pathend_last, * gl_pathv points to (gl_offs + gl_pathc + 1) items. */ static int -globextend(const Char *path, glob_t *pglob, struct glob_limit *limit) +globextend(const Char *path, glob_t *pglob, struct glob_limit *limit, + const char *origpat) { char **pathv; size_t i, newsize, len; @@ -761,7 +856,7 @@ globextend(const Char *path, glob_t *pglob, struct glob_limit *limit) if ((pglob->gl_flags & GLOB_LIMIT) && pglob->gl_matchc > limit->l_path_lim) { - errno = 0; + errno = E2BIG; return (GLOB_NOSPACE); } @@ -779,18 +874,26 @@ globextend(const Char *path, glob_t *pglob, struct glob_limit *limit) } pglob->gl_pathv = pathv; - for (p = path; *p++;) - continue; - len = MB_CUR_MAX * (size_t)(p - path); /* XXX overallocation */ - limit->l_string_cnt += len; - if ((pglob->gl_flags & GLOB_LIMIT) && - limit->l_string_cnt >= GLOB_LIMIT_STRING) { - errno = 0; - return (GLOB_NOSPACE); + if (origpat != NULL) + copy = strdup(origpat); + else { + for (p = path; *p++ != EOS;) + continue; + len = MB_CUR_MAX * (size_t)(p - path); /* XXX overallocation */ + if ((copy = malloc(len)) != NULL) { + if (g_Ctoc(path, copy, len)) { + free(copy); + errno = E2BIG; + return (GLOB_NOSPACE); + } + } } - if ((copy = malloc(len)) != NULL) { - if (g_Ctoc(path, copy, len)) { + if (copy != NULL) { + limit->l_string_cnt += strlen(copy) + 1; + if ((pglob->gl_flags & GLOB_LIMIT) && + limit->l_string_cnt >= GLOB_LIMIT_STRING) { free(copy); + errno = E2BIG; return (GLOB_NOSPACE); } pathv[pglob->gl_offs + pglob->gl_pathc++] = copy; @@ -830,15 +933,17 @@ match(Char *name, Char *pat, Char *patend) ok = 0; if ((k = *name++) == EOS) return (0); - if ((negate_range = ((*pat & M_MASK) == M_NOT)) != EOS) + if ((negate_range = ((*pat & M_MASK) == M_NOT)) != 0) ++pat; while (((c = *pat++) & M_MASK) != M_END) if ((*pat & M_MASK) == M_RNG) { if (table->__collate_load_error ? - CHAR(c) <= CHAR(k) && CHAR(k) <= CHAR(pat[1]) : - __wcollate_range_cmp(CHAR(c), CHAR(k)) <= 0 - && __wcollate_range_cmp(CHAR(k), CHAR(pat[1])) <= 0 - ) + CHAR(c) <= CHAR(k) && + CHAR(k) <= CHAR(pat[1]) : + __wcollate_range_cmp(CHAR(c), + CHAR(k)) <= 0 && + __wcollate_range_cmp(CHAR(k), + CHAR(pat[1])) <= 0) ok = 1; pat += 2; } else if (c == k) @@ -875,13 +980,15 @@ globfree(glob_t *pglob) static DIR * g_opendir(Char *str, glob_t *pglob) { - char buf[MAXPATHLEN]; + char buf[MAXPATHLEN + MB_LEN_MAX - 1]; - if (!*str) + if (*str == EOS) strcpy(buf, "."); else { - if (g_Ctoc(str, buf, sizeof(buf))) + if (g_Ctoc(str, buf, sizeof(buf))) { + errno = ENAMETOOLONG; return (NULL); + } } if (pglob->gl_flags & GLOB_ALTDIRFUNC) @@ -893,7 +1000,7 @@ g_opendir(Char *str, glob_t *pglob) static int g_lstat(Char *fn, struct stat *sb, glob_t *pglob) { - char buf[MAXPATHLEN]; + char buf[MAXPATHLEN + MB_LEN_MAX - 1]; if (g_Ctoc(fn, buf, sizeof(buf))) { errno = ENAMETOOLONG; @@ -907,7 +1014,7 @@ g_lstat(Char *fn, struct stat *sb, glob_t *pglob) static int g_stat(Char *fn, struct stat *sb, glob_t *pglob) { - char buf[MAXPATHLEN]; + char buf[MAXPATHLEN + MB_LEN_MAX - 1]; if (g_Ctoc(fn, buf, sizeof(buf))) { errno = ENAMETOOLONG; @@ -937,10 +1044,14 @@ g_Ctoc(const Char *str, char *buf, size_t len) memset(&mbs, 0, sizeof(mbs)); while (len >= MB_CUR_MAX) { - clen = wcrtomb(buf, *str, &mbs); - if (clen == (size_t)-1) - return (1); - if (*str == L'\0') + clen = wcrtomb(buf, CHAR(*str), &mbs); + if (clen == (size_t)-1) { + /* XXX See initial comment #2. */ + *buf = (char)CHAR(*str); + clen = 1; + memset(&mbs, 0, sizeof(mbs)); + } + if (CHAR(*str) == EOS) return (0); str++; buf += clen; @@ -949,21 +1060,46 @@ g_Ctoc(const Char *str, char *buf, size_t len) return (1); } +static int +err_nomatch(glob_t *pglob, struct glob_limit *limit, const char *origpat) { + /* + * If there was no match we are going to append the origpat + * if GLOB_NOCHECK was specified or if GLOB_NOMAGIC was specified + * and the origpat did not contain any magic characters + * GLOB_NOMAGIC is there just for compatibility with csh. + */ + if ((pglob->gl_flags & GLOB_NOCHECK) || + ((pglob->gl_flags & GLOB_NOMAGIC) && + !(pglob->gl_flags & GLOB_MAGCHAR))) + return (globextend(NULL, pglob, limit, origpat)); + return (GLOB_NOMATCH); +} + +static int +err_aborted(glob_t *pglob, int err, char *buf) { + if ((pglob->gl_errfunc != NULL && pglob->gl_errfunc(buf, err)) || + (pglob->gl_flags & GLOB_ERR)) + return (GLOB_ABORTED); + return (0); +} + #ifdef DEBUG static void qprintf(const char *str, Char *s) { Char *p; - (void)printf("%s:\n", str); - for (p = s; *p; p++) - (void)printf("%c", CHAR(*p)); - (void)printf("\n"); - for (p = s; *p; p++) - (void)printf("%c", *p & M_PROTECT ? '"' : ' '); - (void)printf("\n"); - for (p = s; *p; p++) - (void)printf("%c", ismeta(*p) ? '_' : ' '); - (void)printf("\n"); + (void)printf("%s\n", str); + if (s != NULL) { + for (p = s; *p != EOS; p++) + (void)printf("%c", (char)CHAR(*p)); + (void)printf("\n"); + for (p = s; *p != EOS; p++) + (void)printf("%c", (isprot(*p) ? '\\' : ' ')); + (void)printf("\n"); + for (p = s; *p != EOS; p++) + (void)printf("%c", (ismeta(*p) ? '_' : ' ')); + (void)printf("\n"); + } } #endif diff --git a/lib/libc/sys/ptrace.2 b/lib/libc/sys/ptrace.2 index a94e314..fc5c0ea 100644 --- a/lib/libc/sys/ptrace.2 +++ b/lib/libc/sys/ptrace.2 @@ -2,7 +2,7 @@ .\" $NetBSD: ptrace.2,v 1.2 1995/02/27 12:35:37 cgd Exp $ .\" .\" This file is in the public domain. -.Dd October 20, 2015 +.Dd July 28, 2016 .Dt PTRACE 2 .Os .Sh NAME @@ -58,8 +58,9 @@ The signal may be a normal process signal, generated as a result of traced process behavior, or use of the .Xr kill 2 system call; alternatively, it may be generated by the tracing facility -as a result of attaching, system calls, or stepping by the tracing -process. +as a result of attaching, stepping by the tracing +process, +or an event in the traced process. The tracing process may choose to intercept the signal, using it to observe process behavior (such as .Dv SIGTRAP ) , @@ -69,6 +70,121 @@ The system call is the mechanism by which all this happens. .Pp +A traced process may report additional signal stops corresponding to +events in the traced process. +These additional signal stops are reported as +.Dv SIGTRAP +or +.Dv SIGSTOP +signals. +The tracing process can use the +.Dv PT_LWPINFO +request to determine which events are associated with a +.Dv SIGTRAP +or +.Dv SIGSTOP +signal. +Note that multiple events may be associated with a single signal. +For example, events indicated by the +.Dv PL_FLAG_BORN , +.Dv PL_FLAG_FORKED , +and +.Dv PL_FLAG_EXEC +flags are also reported as a system call exit event +.Pq Dv PL_FLAG_SCX . +The signal stop for a new child process enabled via +.Dv PTRACE_FORK +will report a +.Dv SIGSTOP +signal. +All other additional signal stops use +.Dv SIGTRAP . +.Pp +Each traced process has a tracing event mask. +An event in the traced process only reports a +signal stop if the corresponding flag is set in the tracing event mask. +The current set of tracing event flags include: +.Bl -tag -width ".Dv PTRACE_SYSCALL" +.It Dv PTRACE_EXEC +Report a stop for a successful invocation of +.Xr execve 2 . +This event is indicated by the +.Dv PL_FLAG_EXEC +flag in the +.Va pl_flags +member of +.Vt "struct ptrace_lwpinfo" . +.It Dv PTRACE_SCE +Report a stop on each system call entry. +This event is indicated by the +.Dv PL_FLAG_SCE +flag in the +.Va pl_flags +member of +.Vt "struct ptrace_lwpinfo" . +.It Dv PTRACE_SCX +Report a stop on each system call exit. +This event is indicated by the +.Dv PL_FLAG_SCX +flag in the +.Va pl_flags +member of +.Vt "struct ptrace_lwpinfo" . +.It Dv PTRACE_SYSCALL +Report stops for both system call entry and exit. +.It Dv PTRACE_FORK +This event flag controls tracing for new child processes of a traced process. +.Pp +When this event flag is enabled, +new child processes will enable tracing and stop before executing their +first instruction. +The new child process will include the +.Dv PL_FLAG_CHILD +flag in the +.Va pl_flags +member of +.Vt "struct ptrace_lwpinfo" . +The traced process will report a stop that includes the +.Dv PL_FLAG_FORKED +flag. +The process ID of the new child process will also be present in the +.Va pl_child_pid +member of +.Vt "struct ptrace_lwpinfo" . +Note that new child processes will be attached with the default +tracing event mask; +they do not inherit the event mask of the traced process. +.Pp +When this event flag is not enabled, +new child processes will execute without tracing enabled. +.It Dv PTRACE_LWP +This event flag controls tracing of LWP +.Pq kernel thread +creation and destruction. +When this event is enabled, +new LWPs will stop and report an event with +.Dv PL_FLAG_BORN +set before executing their first instruction, +and exiting LWPs will stop and report an event with +.Dv PL_FLAG_EXITED +set before completing their termination. +.Pp +Note that new processes do not report an event for the creation of their +initial thread, +and exiting processes do not report an event for the termination of the +last thread. +.El +.Pp +The default tracing event mask when attaching to a process via +.Dv PT_ATTACH , +.Dv PT_TRACE_ME , +or +.Dv PTRACE_FORK +includes only +.Dv PTRACE_EXEC +events. +All other event flags are disabled. +.Pp The .Fa request argument specifies what operation is being performed; the meaning of @@ -368,10 +484,23 @@ The process identifier of the new process is available in the member of .Vt "struct ptrace_lwpinfo" . .It PL_FLAG_CHILD -The flag is set for first event reported from a new child, which is -automatically attached due to -.Dv PT_FOLLOW_FORK -enabled. +The flag is set for first event reported from a new child which is +automatically attached when +.Dv PTRACE_FORK +is enabled. +.It PL_FLAG_BORN +This flag is set for the first event reported from a new LWP when +.Dv PTRACE_LWP +is enabled. +It is reported along with +.Dv PL_FLAG_SCX . +.It PL_FLAG_EXITED +This flag is set for the last event reported by an exiting LWP when +.Dv PTRACE_LWP +is enabled. +Note that this event is not reported when the last LWP in a process exits. +The termination of the last thread is reported via a normal process exit +event. .El .It pl_sigmask The current signal mask of the LWP @@ -442,27 +571,72 @@ This request will suspend the specified thread. .It PT_RESUME This request will resume the specified thread. .It PT_TO_SCE -This request will trace the specified process on each system call entry. +This request will set the +.Dv PTRACE_SCE +event flag to trace all future system call entries and continue the process. +The +.Fa addr +and +.Fa data +arguments are used the same as for +.Dv PT_CONTINUE. .It PT_TO_SCX -This request will trace the specified process on each system call exit. +This request will set the +.Dv PTRACE_SCX +event flag to trace all future system call exits and continue the process. +The +.Fa addr +and +.Fa data +arguments are used the same as for +.Dv PT_CONTINUE. .It PT_SYSCALL -This request will trace the specified process -on each system call entry and exit. +This request will set the +.Dv PTRACE_SYSCALL +event flag to trace all future system call entries and exits and continue +the process. +The +.Fa addr +and +.Fa data +arguments are used the same as for +.Dv PT_CONTINUE. .It PT_FOLLOW_FORK This request controls tracing for new child processes of a traced process. If .Fa data is non-zero, -then new child processes will enable tracing and stop before executing their -first instruction. +.Dv PTRACE_FORK +is set in the traced process's event tracing mask. If .Fa data -is zero, then new child processes will execute without tracing enabled. -By default, tracing is not enabled for new child processes. -Child processes do not inherit this property. -The traced process will set the -.Dv PL_FLAG_FORKED -flag upon exit from a system call that creates a new process. +is zero, +.Dv PTRACE_FORK +is cleared from the traced process's event tracing mask. +.It PT_LWP_EVENTS +This request controls tracing of LWP creation and destruction. +If +.Fa data +is non-zero, +.Dv PTRACE_LWP +is set in the traced process's event tracing mask. +If +.Fa data +is zero, +.Dv PTRACE_LWP +is cleared from the traced process's event tracing mask. +.It PT_GET_EVENT_MASK +This request reads the traced process's event tracing mask into the +integer pointed to by +.Fa addr . +The size of the integer must be passed in +.Fa data . +.It PT_SET_EVENT_MASK +This request sets the traced process's event tracing mask from the +integer pointed to by +.Fa addr . +The size of the integer must be passed in +.Fa data . .It PT_VM_TIMESTAMP This request returns the generation number or timestamp of the memory map of the traced process as the return value from diff --git a/release/doc/share/xml/errata.xml b/release/doc/share/xml/errata.xml index f5c93bc..a1453e9 100644 --- a/release/doc/share/xml/errata.xml +++ b/release/doc/share/xml/errata.xml @@ -48,6 +48,62 @@ <entry><para>Fix &man.freebsd-update.8; support of &os; 11.0-RELEASE</para></entry> </row> + + <row> + <entry><link + xlink:href="&security.url;/FreeBSD-EN-16:10.dhclient.asc">FreeBSD-EN-16:10.dhclient</link></entry> + <entry>11 August 2016</entry> + <entry><para>Better handle unknown options received from + aDHCP server</para></entry> + </row> + + <row> + <entry><link + xlink:href="&security.url;/FreeBSD-EN-16:11.vmbus.asc">FreeBSD-EN-16:11.vmbus</link></entry> + <entry>11 August 2016</entry> + <entry><para>Avoid using spin locks for channel message + locks</para></entry> + </row> + + <row> + <entry><link + xlink:href="&security.url;/FreeBSD-EN-16:12.hv_storvsc.asc">FreeBSD-EN-16:12.hv_storvsc</link></entry> + <entry>11 August 2016</entry> + <entry><para>Enable INQUIRY result check only on + Windows 10 host systems</para></entry> + </row> + + <row> + <entry><link + xlink:href="&security.url;/FreeBSD-EN-16:13.vmbus.asc">FreeBSD-EN-16:13.vmbus</link></entry> + <entry>11 August 2016</entry> + <entry><para>Register time counter early enough for TSC freq + calibration</para></entry> + </row> + + <row> + <entry><link + xlink:href="&security.url;/FreeBSD-EN-16:14.hv_storvsc.asc">FreeBSD-EN-16:14.hv_storvsc</link></entry> + <entry>11 August 2016</entry> + <entry><para>Disable incorrect callout in + &man.hv.storvsc.4;</para></entry> + </row> + + <row> + <entry><link + xlink:href="&security.url;/FreeBSD-EN-16:15.vmbus.asc">FreeBSD-EN-16:15.vmbus</link></entry> + <entry>11 August 2016</entry> + <entry><para>Better handle the GPADL setup failure in + Hyper-V</para></entry> + </row> + + <row> + <entry><link + xlink:href="&security.url;/FreeBSD-EN-16:16.hv_storvsc.asc">FreeBSD-EN-16:16.hv_storvsc</link></entry> + <entry>11 August 2016</entry> + <entry><para>Fix SCSI INQUIRY checks and error + handling</para></entry> + </row> </tbody> </tgroup> </informaltable> diff --git a/release/release.sh b/release/release.sh index caa6f26..6dc4a28 100755 --- a/release/release.sh +++ b/release/release.sh @@ -281,9 +281,15 @@ extra_chroot_setup() { fi if [ ! -z "${EMBEDDEDPORTS}" ]; then + _OSVERSION=$(chroot ${CHROOTDIR} /usr/bin/uname -U) + REVISION=$(chroot ${CHROOTDIR} make -C /usr/src/release -V REVISION) + BRANCH=$(chroot ${CHROOTDIR} make -C /usr/src/release -V BRANCH) + PBUILD_FLAGS="OSVERSION=${_OSVERSION} BATCH=yes" + PBUILD_FLAGS="${PBUILD_FLAGS} UNAME_r=${UNAME_r}" + PBUILD_FLAGS="${PBUILD_FLAGS} OSREL=${REVISION}" for _PORT in ${EMBEDDEDPORTS}; do eval chroot ${CHROOTDIR} make -C /usr/ports/${_PORT} \ - BATCH=1 FORCE_PKG_REGISTER=1 install clean distclean + FORCE_PKG_REGISTER=1 ${PBUILD_FLAGS} install clean distclean done fi diff --git a/sbin/pfctl/parse.y b/sbin/pfctl/parse.y index 16c7847..5c0becd 100644 --- a/sbin/pfctl/parse.y +++ b/sbin/pfctl/parse.y @@ -3718,8 +3718,8 @@ tos : STRING { else if ($1[0] == '0' && $1[1] == 'x') $$ = strtoul($1, NULL, 16); else - $$ = 0; /* flag bad argument */ - if (!$$ || $$ > 255) { + $$ = 256; /* flag bad argument */ + if ($$ < 0 || $$ > 255) { yyerror("illegal tos value %s", $1); free($1); YYERROR; @@ -3728,7 +3728,7 @@ tos : STRING { } | NUMBER { $$ = $1; - if (!$$ || $$ > 255) { + if ($$ < 0 || $$ > 255) { yyerror("illegal tos value %s", $1); YYERROR; } diff --git a/share/mk/bsd.README b/share/mk/bsd.README index 6872a6b..df973d5 100644 --- a/share/mk/bsd.README +++ b/share/mk/bsd.README @@ -272,6 +272,7 @@ PROGS_CXX PROG and PROGS_CXX in one Makefile. To define - DEBUG_FLAGS - DPADD - DPSRCS + - INTERNALPROG (no installation) - LDADD - LDFLAGS - LINKS diff --git a/share/mk/bsd.progs.mk b/share/mk/bsd.progs.mk index 69cb1bf..4bf5324 100644 --- a/share/mk/bsd.progs.mk +++ b/share/mk/bsd.progs.mk @@ -35,8 +35,8 @@ UPDATE_DEPENDFILE_PROG?= no # just one of many PROG_OVERRIDE_VARS += BINDIR BINGRP BINOWN BINMODE DPSRCS MAN NO_WERROR \ PROGNAME SRCS STRIP WARNS -PROG_VARS += CFLAGS CXXFLAGS DEBUG_FLAGS DPADD LDADD LIBADD LINKS \ - LDFLAGS MLINKS ${PROG_OVERRIDE_VARS} +PROG_VARS += CFLAGS CXXFLAGS DEBUG_FLAGS DPADD INTERNALPROG LDADD LIBADD \ + LINKS LDFLAGS MLINKS ${PROG_OVERRIDE_VARS} .for v in ${PROG_VARS:O:u} .if empty(${PROG_OVERRIDE_VARS:M$v}) .if defined(${v}.${PROG}) diff --git a/sys/amd64/amd64/initcpu.c b/sys/amd64/amd64/initcpu.c index 8868cd3..e20a271 100644 --- a/sys/amd64/amd64/initcpu.c +++ b/sys/amd64/amd64/initcpu.c @@ -125,6 +125,20 @@ init_amd(void) wrmsr(MSR_NB_CFG1, msr); } } + + /* + * BIOS may configure Family 10h processors to convert WC+ cache type + * to CD. That can hurt performance of guest VMs using nested paging. + * The relevant MSR bit is not documented in the BKDG, + * the fix is borrowed from Linux. + */ + if (CPUID_TO_FAMILY(cpu_id) == 0x10) { + if ((cpu_feature2 & CPUID2_HV) == 0) { + msr = rdmsr(0xc001102a); + msr &= ~((uint64_t)1 << 24); + wrmsr(0xc001102a, msr); + } + } } /* diff --git a/sys/amd64/amd64/trap.c b/sys/amd64/amd64/trap.c index 218c8cf..f487c44 100644 --- a/sys/amd64/amd64/trap.c +++ b/sys/amd64/amd64/trap.c @@ -448,8 +448,8 @@ trap(struct trapframe *frame) goto out; case T_DNA: - KASSERT(!PCB_USER_FPU(td->td_pcb), - ("Unregistered use of FPU in kernel")); + if (PCB_USER_FPU(td->td_pcb)) + panic("Unregistered use of FPU in kernel"); fpudna(); goto out; diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c index 9430037..1700194 100644 --- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c +++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/arc.c @@ -21,7 +21,7 @@ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. * Copyright (c) 2012, Joyent, Inc. All rights reserved. - * Copyright (c) 2011, 2015 by Delphix. All rights reserved. + * Copyright (c) 2011, 2016 by Delphix. All rights reserved. * Copyright (c) 2014 by Saso Kiselkov. All rights reserved. * Copyright 2015 Nexenta Systems, Inc. All rights reserved. */ @@ -782,6 +782,7 @@ typedef struct arc_write_callback arc_write_callback_t; struct arc_write_callback { void *awcb_private; arc_done_func_t *awcb_ready; + arc_done_func_t *awcb_children_ready; arc_done_func_t *awcb_physdone; arc_done_func_t *awcb_done; arc_buf_t *awcb_buf; @@ -5045,6 +5046,15 @@ arc_write_ready(zio_t *zio) hdr->b_flags |= ARC_FLAG_IO_IN_PROGRESS; } +static void +arc_write_children_ready(zio_t *zio) +{ + arc_write_callback_t *callback = zio->io_private; + arc_buf_t *buf = callback->awcb_buf; + + callback->awcb_children_ready(zio, buf, callback->awcb_private); +} + /* * The SPA calls this callback for each physical write that happens on behalf * of a logical write. See the comment in dbuf_write_physdone() for details. @@ -5141,7 +5151,8 @@ arc_write_done(zio_t *zio) zio_t * arc_write(zio_t *pio, spa_t *spa, uint64_t txg, blkptr_t *bp, arc_buf_t *buf, boolean_t l2arc, boolean_t l2arc_compress, - const zio_prop_t *zp, arc_done_func_t *ready, arc_done_func_t *physdone, + const zio_prop_t *zp, arc_done_func_t *ready, + arc_done_func_t *children_ready, arc_done_func_t *physdone, arc_done_func_t *done, void *private, zio_priority_t priority, int zio_flags, const zbookmark_phys_t *zb) { @@ -5161,13 +5172,16 @@ arc_write(zio_t *pio, spa_t *spa, uint64_t txg, hdr->b_flags |= ARC_FLAG_L2COMPRESS; callback = kmem_zalloc(sizeof (arc_write_callback_t), KM_SLEEP); callback->awcb_ready = ready; + callback->awcb_children_ready = children_ready; callback->awcb_physdone = physdone; callback->awcb_done = done; callback->awcb_private = private; callback->awcb_buf = buf; zio = zio_write(pio, spa, txg, bp, buf->b_data, hdr->b_size, zp, - arc_write_ready, arc_write_physdone, arc_write_done, callback, + arc_write_ready, + (children_ready != NULL) ? arc_write_children_ready : NULL, + arc_write_physdone, arc_write_done, callback, priority, zio_flags, zb); return (zio); diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dbuf.c b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dbuf.c index 2f26e9fb..9d5c398 100644 --- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dbuf.c +++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dbuf.c @@ -481,13 +481,49 @@ dbuf_verify(dmu_buf_impl_t *db) * If the blkptr isn't set but they have nonzero data, * it had better be dirty, otherwise we'll lose that * data when we evict this buffer. + * + * There is an exception to this rule for indirect blocks; in + * this case, if the indirect block is a hole, we fill in a few + * fields on each of the child blocks (importantly, birth time) + * to prevent hole birth times from being lost when you + * partially fill in a hole. */ if (db->db_dirtycnt == 0) { - uint64_t *buf = db->db.db_data; - int i; + if (db->db_level == 0) { + uint64_t *buf = db->db.db_data; + int i; - for (i = 0; i < db->db.db_size >> 3; i++) { - ASSERT(buf[i] == 0); + for (i = 0; i < db->db.db_size >> 3; i++) { + ASSERT(buf[i] == 0); + } + } else { + blkptr_t *bps = db->db.db_data; + ASSERT3U(1 << DB_DNODE(db)->dn_indblkshift, ==, + db->db.db_size); + /* + * We want to verify that all the blkptrs in the + * indirect block are holes, but we may have + * automatically set up a few fields for them. + * We iterate through each blkptr and verify + * they only have those fields set. + */ + for (int i = 0; + i < db->db.db_size / sizeof (blkptr_t); + i++) { + blkptr_t *bp = &bps[i]; + ASSERT(ZIO_CHECKSUM_IS_ZERO( + &bp->blk_cksum)); + ASSERT( + DVA_IS_EMPTY(&bp->blk_dva[0]) && + DVA_IS_EMPTY(&bp->blk_dva[1]) && + DVA_IS_EMPTY(&bp->blk_dva[2])); + ASSERT0(bp->blk_fill); + ASSERT0(bp->blk_pad[0]); + ASSERT0(bp->blk_pad[1]); + ASSERT(!BP_IS_EMBEDDED(bp)); + ASSERT(BP_IS_HOLE(bp)); + ASSERT0(bp->blk_phys_birth); + } } } } @@ -655,10 +691,31 @@ dbuf_read_impl(dmu_buf_impl_t *db, zio_t *zio, uint32_t flags) BP_IS_HOLE(db->db_blkptr)))) { arc_buf_contents_t type = DBUF_GET_BUFC_TYPE(db); - DB_DNODE_EXIT(db); dbuf_set_data(db, arc_buf_alloc(db->db_objset->os_spa, db->db.db_size, db, type)); bzero(db->db.db_data, db->db.db_size); + + if (db->db_blkptr != NULL && db->db_level > 0 && + BP_IS_HOLE(db->db_blkptr) && + db->db_blkptr->blk_birth != 0) { + blkptr_t *bps = db->db.db_data; + for (int i = 0; i < ((1 << + DB_DNODE(db)->dn_indblkshift) / sizeof (blkptr_t)); + i++) { + blkptr_t *bp = &bps[i]; + ASSERT3U(BP_GET_LSIZE(db->db_blkptr), ==, + 1 << dn->dn_indblkshift); + BP_SET_LSIZE(bp, + BP_GET_LEVEL(db->db_blkptr) == 1 ? + dn->dn_datablksz : + BP_GET_LSIZE(db->db_blkptr)); + BP_SET_TYPE(bp, BP_GET_TYPE(db->db_blkptr)); + BP_SET_LEVEL(bp, + BP_GET_LEVEL(db->db_blkptr) - 1); + BP_SET_BIRTH(bp, db->db_blkptr->blk_birth, 0); + } + } + DB_DNODE_EXIT(db); db->db_state = DB_CACHED; mutex_exit(&db->db_mtx); return; @@ -2876,7 +2933,8 @@ dbuf_write_ready(zio_t *zio, arc_buf_t *buf, void *vdb) uint64_t fill = 0; int i; - ASSERT3P(db->db_blkptr, ==, bp); + ASSERT3P(db->db_blkptr, !=, NULL); + ASSERT3P(&db->db_data_pending->dr_bp_copy, ==, bp); DB_DNODE_ENTER(db); dn = DB_DNODE(db); @@ -2898,7 +2956,7 @@ dbuf_write_ready(zio_t *zio, arc_buf_t *buf, void *vdb) #ifdef ZFS_DEBUG if (db->db_blkid == DMU_SPILL_BLKID) { ASSERT(dn->dn_phys->dn_flags & DNODE_FLAG_SPILL_BLKPTR); - ASSERT(!(BP_IS_HOLE(db->db_blkptr)) && + ASSERT(!(BP_IS_HOLE(bp)) && db->db_blkptr == &dn->dn_phys->dn_spill); } #endif @@ -2939,6 +2997,49 @@ dbuf_write_ready(zio_t *zio, arc_buf_t *buf, void *vdb) bp->blk_fill = fill; mutex_exit(&db->db_mtx); + + rw_enter(&dn->dn_struct_rwlock, RW_WRITER); + *db->db_blkptr = *bp; + rw_exit(&dn->dn_struct_rwlock); +} + +/* ARGSUSED */ +/* + * This function gets called just prior to running through the compression + * stage of the zio pipeline. If we're an indirect block comprised of only + * holes, then we want this indirect to be compressed away to a hole. In + * order to do that we must zero out any information about the holes that + * this indirect points to prior to before we try to compress it. + */ +static void +dbuf_write_children_ready(zio_t *zio, arc_buf_t *buf, void *vdb) +{ + dmu_buf_impl_t *db = vdb; + dnode_t *dn; + blkptr_t *bp; + uint64_t i; + int epbs; + + ASSERT3U(db->db_level, >, 0); + DB_DNODE_ENTER(db); + dn = DB_DNODE(db); + epbs = dn->dn_phys->dn_indblkshift - SPA_BLKPTRSHIFT; + + /* Determine if all our children are holes */ + for (i = 0, bp = db->db.db_data; i < 1 << epbs; i++, bp++) { + if (!BP_IS_HOLE(bp)) + break; + } + + /* + * If all the children are holes, then zero them all out so that + * we may get compressed away. + */ + if (i == 1 << epbs) { + /* didn't find any non-holes */ + bzero(db->db.db_data, db->db.db_size); + } + DB_DNODE_EXIT(db); } /* @@ -3117,6 +3218,8 @@ dbuf_write(dbuf_dirty_record_t *dr, arc_buf_t *data, dmu_tx_t *tx) zio_t *zio; int wp_flag = 0; + ASSERT(dmu_tx_is_syncing(tx)); + DB_DNODE_ENTER(db); dn = DB_DNODE(db); os = dn->dn_objset; @@ -3175,6 +3278,14 @@ dbuf_write(dbuf_dirty_record_t *dr, arc_buf_t *data, dmu_tx_t *tx) dmu_write_policy(os, dn, db->db_level, wp_flag, &zp); DB_DNODE_EXIT(db); + /* + * We copy the blkptr now (rather than when we instantiate the dirty + * record), because its value can change between open context and + * syncing context. We do not need to hold dn_struct_rwlock to read + * db_blkptr because we are in syncing context. + */ + dr->dr_bp_copy = *db->db_blkptr; + if (db->db_level == 0 && dr->dt.dl.dr_override_state == DR_OVERRIDDEN) { /* @@ -3184,8 +3295,9 @@ dbuf_write(dbuf_dirty_record_t *dr, arc_buf_t *data, dmu_tx_t *tx) void *contents = (data != NULL) ? data->b_data : NULL; dr->dr_zio = zio_write(zio, os->os_spa, txg, - db->db_blkptr, contents, db->db.db_size, &zp, - dbuf_write_override_ready, NULL, dbuf_write_override_done, + &dr->dr_bp_copy, contents, db->db.db_size, &zp, + dbuf_write_override_ready, NULL, NULL, + dbuf_write_override_done, dr, ZIO_PRIORITY_ASYNC_WRITE, ZIO_FLAG_MUSTSUCCEED, &zb); mutex_enter(&db->db_mtx); dr->dt.dl.dr_override_state = DR_NOT_OVERRIDDEN; @@ -3196,15 +3308,27 @@ dbuf_write(dbuf_dirty_record_t *dr, arc_buf_t *data, dmu_tx_t *tx) ASSERT(zp.zp_checksum == ZIO_CHECKSUM_OFF || zp.zp_checksum == ZIO_CHECKSUM_NOPARITY); dr->dr_zio = zio_write(zio, os->os_spa, txg, - db->db_blkptr, NULL, db->db.db_size, &zp, - dbuf_write_nofill_ready, NULL, dbuf_write_nofill_done, db, + &dr->dr_bp_copy, NULL, db->db.db_size, &zp, + dbuf_write_nofill_ready, NULL, NULL, + dbuf_write_nofill_done, db, ZIO_PRIORITY_ASYNC_WRITE, ZIO_FLAG_MUSTSUCCEED | ZIO_FLAG_NODATA, &zb); } else { ASSERT(arc_released(data)); + + /* + * For indirect blocks, we want to setup the children + * ready callback so that we can properly handle an indirect + * block that only contains holes. + */ + arc_done_func_t *children_ready_cb = NULL; + if (db->db_level != 0) + children_ready_cb = dbuf_write_children_ready; + dr->dr_zio = arc_write(zio, os->os_spa, txg, - db->db_blkptr, data, DBUF_IS_L2CACHEABLE(db), + &dr->dr_bp_copy, data, DBUF_IS_L2CACHEABLE(db), DBUF_IS_L2COMPRESSIBLE(db), &zp, dbuf_write_ready, + children_ready_cb, dbuf_write_physdone, dbuf_write_done, db, ZIO_PRIORITY_ASYNC_WRITE, ZIO_FLAG_MUSTSUCCEED, &zb); } diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu.c b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu.c index d8d186c..9ce9665 100644 --- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu.c +++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu.c @@ -20,7 +20,7 @@ */ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. - * Copyright (c) 2011, 2015 by Delphix. All rights reserved. + * Copyright (c) 2011, 2016 by Delphix. All rights reserved. */ /* Copyright (c) 2013 by Saso Kiselkov. All rights reserved. */ /* Copyright (c) 2013, Joyent, Inc. All rights reserved. */ @@ -1590,10 +1590,11 @@ dmu_sync_late_arrival(zio_t *pio, objset_t *os, dmu_sync_cb_t *done, zgd_t *zgd, dsa->dsa_zgd = zgd; dsa->dsa_tx = tx; - zio_nowait(zio_write(pio, os->os_spa, dmu_tx_get_txg(tx), zgd->zgd_bp, - zgd->zgd_db->db_data, zgd->zgd_db->db_size, zp, - dmu_sync_late_arrival_ready, NULL, dmu_sync_late_arrival_done, dsa, - ZIO_PRIORITY_SYNC_WRITE, ZIO_FLAG_CANFAIL, zb)); + zio_nowait(zio_write(pio, os->os_spa, dmu_tx_get_txg(tx), + zgd->zgd_bp, zgd->zgd_db->db_data, zgd->zgd_db->db_size, + zp, dmu_sync_late_arrival_ready, NULL, + NULL, dmu_sync_late_arrival_done, dsa, ZIO_PRIORITY_SYNC_WRITE, + ZIO_FLAG_CANFAIL, zb)); return (0); } @@ -1746,8 +1747,8 @@ dmu_sync(zio_t *pio, uint64_t txg, dmu_sync_cb_t *done, zgd_t *zgd) zio_nowait(arc_write(pio, os->os_spa, txg, bp, dr->dt.dl.dr_data, DBUF_IS_L2CACHEABLE(db), DBUF_IS_L2COMPRESSIBLE(db), &zp, dmu_sync_ready, - NULL, dmu_sync_done, dsa, ZIO_PRIORITY_SYNC_WRITE, - ZIO_FLAG_CANFAIL, &zb)); + NULL, NULL, dmu_sync_done, dsa, + ZIO_PRIORITY_SYNC_WRITE, ZIO_FLAG_CANFAIL, &zb)); return (0); } diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_objset.c b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_objset.c index 367dbcb..82c5854 100644 --- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_objset.c +++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dmu_objset.c @@ -20,7 +20,7 @@ */ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. - * Copyright (c) 2012, 2014 by Delphix. All rights reserved. + * Copyright (c) 2012, 2016 by Delphix. All rights reserved. * Copyright (c) 2013 by Saso Kiselkov. All rights reserved. * Copyright (c) 2013, Joyent, Inc. All rights reserved. * Copyright (c) 2014 Spectra Logic Corporation, All rights reserved. @@ -1116,9 +1116,9 @@ dmu_objset_sync(objset_t *os, zio_t *pio, dmu_tx_t *tx) zio = arc_write(pio, os->os_spa, tx->tx_txg, os->os_rootbp, os->os_phys_buf, DMU_OS_IS_L2CACHEABLE(os), - DMU_OS_IS_L2COMPRESSIBLE(os), &zp, dmu_objset_write_ready, - NULL, dmu_objset_write_done, os, ZIO_PRIORITY_ASYNC_WRITE, - ZIO_FLAG_MUSTSUCCEED, &zb); + DMU_OS_IS_L2COMPRESSIBLE(os), + &zp, dmu_objset_write_ready, NULL, NULL, dmu_objset_write_done, + os, ZIO_PRIORITY_ASYNC_WRITE, ZIO_FLAG_MUSTSUCCEED, &zb); /* * Sync special dnodes - the parent IO for the sync is the root block diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dnode_sync.c b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dnode_sync.c index 9aee513..7179c41 100644 --- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dnode_sync.c +++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dnode_sync.c @@ -21,7 +21,7 @@ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. - * Copyright (c) 2012, 2015 by Delphix. All rights reserved. + * Copyright (c) 2012, 2016 by Delphix. All rights reserved. * Copyright (c) 2014 Spectra Logic Corporation, All rights reserved. */ @@ -60,20 +60,14 @@ dnode_increase_indirection(dnode_t *dn, dmu_tx_t *tx) dprintf("os=%p obj=%llu, increase to %d\n", dn->dn_objset, dn->dn_object, dn->dn_phys->dn_nlevels); - /* check for existing blkptrs in the dnode */ - for (i = 0; i < nblkptr; i++) - if (!BP_IS_HOLE(&dn->dn_phys->dn_blkptr[i])) - break; - if (i != nblkptr) { - /* transfer dnode's block pointers to new indirect block */ - (void) dbuf_read(db, NULL, DB_RF_MUST_SUCCEED|DB_RF_HAVESTRUCT); - ASSERT(db->db.db_data); - ASSERT(arc_released(db->db_buf)); - ASSERT3U(sizeof (blkptr_t) * nblkptr, <=, db->db.db_size); - bcopy(dn->dn_phys->dn_blkptr, db->db.db_data, - sizeof (blkptr_t) * nblkptr); - arc_buf_freeze(db->db_buf); - } + /* transfer dnode's block pointers to new indirect block */ + (void) dbuf_read(db, NULL, DB_RF_MUST_SUCCEED|DB_RF_HAVESTRUCT); + ASSERT(db->db.db_data); + ASSERT(arc_released(db->db_buf)); + ASSERT3U(sizeof (blkptr_t) * nblkptr, <=, db->db.db_size); + bcopy(dn->dn_phys->dn_blkptr, db->db.db_data, + sizeof (blkptr_t) * nblkptr); + arc_buf_freeze(db->db_buf); /* set dbuf's parent pointers to new indirect buf */ for (i = 0; i < nblkptr; i++) { diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dsl_scan.c b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dsl_scan.c index a157dfe..9880de9 100644 --- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dsl_scan.c +++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/dsl_scan.c @@ -56,7 +56,8 @@ typedef int (scan_cb_t)(dsl_pool_t *, const blkptr_t *, static scan_cb_t dsl_scan_scrub_cb; static void dsl_scan_cancel_sync(void *, dmu_tx_t *); -static void dsl_scan_sync_state(dsl_scan_t *, dmu_tx_t *tx); +static void dsl_scan_sync_state(dsl_scan_t *, dmu_tx_t *); +static boolean_t dsl_scan_restarting(dsl_scan_t *, dmu_tx_t *); unsigned int zfs_top_maxinflight = 32; /* maximum I/Os per top-level */ unsigned int zfs_resilver_delay = 2; /* number of ticks to delay resilver */ @@ -329,8 +330,15 @@ dsl_scan_done(dsl_scan_t *scn, boolean_t complete, dmu_tx_t *tx) else scn->scn_phys.scn_state = DSS_CANCELED; - spa_history_log_internal(spa, "scan done", tx, - "complete=%u", complete); + if (dsl_scan_restarting(scn, tx)) + spa_history_log_internal(spa, "scan aborted, restarting", tx, + "errors=%llu", spa_get_errlog_size(spa)); + else if (!complete) + spa_history_log_internal(spa, "scan cancelled", tx, + "errors=%llu", spa_get_errlog_size(spa)); + else + spa_history_log_internal(spa, "scan done", tx, + "errors=%llu", spa_get_errlog_size(spa)); if (DSL_SCAN_IS_SCRUB_RESILVER(scn)) { mutex_enter(&spa->spa_scrub_lock); @@ -1485,8 +1493,7 @@ dsl_scan_sync(dsl_pool_t *dp, dmu_tx_t *tx) * that we can restart an old-style scan while the pool is being * imported (see dsl_scan_init). */ - if (scn->scn_restart_txg != 0 && - scn->scn_restart_txg <= tx->tx_txg) { + if (dsl_scan_restarting(scn, tx)) { pool_scan_func_t func = POOL_SCAN_SCRUB; dsl_scan_done(scn, B_FALSE, tx); if (vdev_resilver_needed(spa->spa_root_vdev, NULL, NULL)) @@ -1913,3 +1920,10 @@ dsl_scan(dsl_pool_t *dp, pool_scan_func_t func) return (dsl_sync_task(spa_name(spa), dsl_scan_setup_check, dsl_scan_setup_sync, &func, 0, ZFS_SPACE_CHECK_NONE)); } + +static boolean_t +dsl_scan_restarting(dsl_scan_t *scn, dmu_tx_t *tx) +{ + return (scn->scn_restart_txg != 0 && + scn->scn_restart_txg <= tx->tx_txg); +} diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/arc.h b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/arc.h index 04a80f7..714c528 100644 --- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/arc.h +++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/arc.h @@ -20,7 +20,7 @@ */ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. - * Copyright (c) 2012, 2014 by Delphix. All rights reserved. + * Copyright (c) 2012, 2016 by Delphix. All rights reserved. * Copyright (c) 2013 by Saso Kiselkov. All rights reserved. */ @@ -141,9 +141,11 @@ int arc_read(zio_t *pio, spa_t *spa, const blkptr_t *bp, arc_flags_t *arc_flags, const zbookmark_phys_t *zb); zio_t *arc_write(zio_t *pio, spa_t *spa, uint64_t txg, blkptr_t *bp, arc_buf_t *buf, boolean_t l2arc, boolean_t l2arc_compress, - const zio_prop_t *zp, arc_done_func_t *ready, arc_done_func_t *physdone, - arc_done_func_t *done, void *priv, zio_priority_t priority, - int zio_flags, const zbookmark_phys_t *zb); + const zio_prop_t *zp, + arc_done_func_t *ready, arc_done_func_t *child_ready, + arc_done_func_t *physdone, arc_done_func_t *done, + void *priv, zio_priority_t priority, int zio_flags, + const zbookmark_phys_t *zb); void arc_freed(spa_t *spa, const blkptr_t *bp); void arc_set_callback(arc_buf_t *buf, arc_evict_func_t *func, void *priv); diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/dbuf.h b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/dbuf.h index 233d541..4964126 100644 --- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/dbuf.h +++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/dbuf.h @@ -121,6 +121,9 @@ typedef struct dbuf_dirty_record { /* How much space was changed to dsl_pool_dirty_space() for this? */ unsigned int dr_accounted; + /* A copy of the bp that points to us */ + blkptr_t dr_bp_copy; + union dirty_types { struct dirty_indirect { diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/zio.h b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/zio.h index 56c821a..fc7be5b 100644 --- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/zio.h +++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/sys/zio.h @@ -22,7 +22,7 @@ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. * Copyright 2011 Nexenta Systems, Inc. All rights reserved. - * Copyright (c) 2012, 2014 by Delphix. All rights reserved. + * Copyright (c) 2012, 2016 by Delphix. All rights reserved. * Copyright (c) 2013 by Saso Kiselkov. All rights reserved. */ @@ -436,6 +436,7 @@ struct zio { /* Callback info */ zio_done_func_t *io_ready; + zio_done_func_t *io_children_ready; zio_done_func_t *io_physdone; zio_done_func_t *io_done; void *io_private; @@ -503,9 +504,10 @@ extern zio_t *zio_read(zio_t *pio, spa_t *spa, const blkptr_t *bp, void *data, extern zio_t *zio_write(zio_t *pio, spa_t *spa, uint64_t txg, blkptr_t *bp, void *data, uint64_t size, const zio_prop_t *zp, - zio_done_func_t *ready, zio_done_func_t *physdone, zio_done_func_t *done, - void *priv, - zio_priority_t priority, enum zio_flag flags, const zbookmark_phys_t *zb); + zio_done_func_t *ready, zio_done_func_t *children_ready, + zio_done_func_t *physdone, zio_done_func_t *done, + void *priv, zio_priority_t priority, enum zio_flag flags, + const zbookmark_phys_t *zb); extern zio_t *zio_rewrite(zio_t *pio, spa_t *spa, uint64_t txg, blkptr_t *bp, void *data, uint64_t size, zio_done_func_t *done, void *priv, diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_vfsops.c b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_vfsops.c index 9bb6091..8523bc4 100644 --- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_vfsops.c +++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_vfsops.c @@ -22,7 +22,7 @@ * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. * Copyright (c) 2011 Pawel Jakub Dawidek <pawel@dawidek.net>. * All rights reserved. - * Copyright (c) 2012, 2014 by Delphix. All rights reserved. + * Copyright (c) 2012, 2015 by Delphix. All rights reserved. * Copyright (c) 2014 Integros [integros.com] */ @@ -847,72 +847,46 @@ zfs_owner_overquota(zfsvfs_t *zfsvfs, znode_t *zp, boolean_t isgroup) return (zfs_fuid_overquota(zfsvfs, isgroup, fuid)); } -int -zfsvfs_create(const char *osname, zfsvfs_t **zfvp) +/* + * Associate this zfsvfs with the given objset, which must be owned. + * This will cache a bunch of on-disk state from the objset in the + * zfsvfs. + */ +static int +zfsvfs_init(zfsvfs_t *zfsvfs, objset_t *os) { - objset_t *os; - zfsvfs_t *zfsvfs; - uint64_t zval; - int i, error; - uint64_t sa_obj; - - /* - * XXX: Fix struct statfs so this isn't necessary! - * - * The 'osname' is used as the filesystem's special node, which means - * it must fit in statfs.f_mntfromname, or else it can't be - * enumerated, so libzfs_mnttab_find() returns NULL, which causes - * 'zfs unmount' to think it's not mounted when it is. - */ - if (strlen(osname) >= MNAMELEN) - return (SET_ERROR(ENAMETOOLONG)); - - zfsvfs = kmem_zalloc(sizeof (zfsvfs_t), KM_SLEEP); - - /* - * We claim to always be readonly so we can open snapshots; - * other ZPL code will prevent us from writing to snapshots. - */ - error = dmu_objset_own(osname, DMU_OST_ZFS, B_TRUE, zfsvfs, &os); - if (error) { - kmem_free(zfsvfs, sizeof (zfsvfs_t)); - return (error); - } + int error; + uint64_t val; - /* - * Initialize the zfs-specific filesystem structure. - * Should probably make this a kmem cache, shuffle fields, - * and just bzero up to z_hold_mtx[]. - */ - zfsvfs->z_vfs = NULL; - zfsvfs->z_parent = zfsvfs; zfsvfs->z_max_blksz = SPA_OLD_MAXBLOCKSIZE; zfsvfs->z_show_ctldir = ZFS_SNAPDIR_VISIBLE; zfsvfs->z_os = os; error = zfs_get_zplprop(os, ZFS_PROP_VERSION, &zfsvfs->z_version); - if (error) { - goto out; - } else if (zfsvfs->z_version > + if (error != 0) + return (error); + if (zfsvfs->z_version > zfs_zpl_version_map(spa_version(dmu_objset_spa(os)))) { (void) printf("Can't mount a version %lld file system " "on a version %lld pool\n. Pool must be upgraded to mount " "this file system.", (u_longlong_t)zfsvfs->z_version, (u_longlong_t)spa_version(dmu_objset_spa(os))); - error = SET_ERROR(ENOTSUP); - goto out; + return (SET_ERROR(ENOTSUP)); } - if ((error = zfs_get_zplprop(os, ZFS_PROP_NORMALIZE, &zval)) != 0) - goto out; - zfsvfs->z_norm = (int)zval; + error = zfs_get_zplprop(os, ZFS_PROP_NORMALIZE, &val); + if (error != 0) + return (error); + zfsvfs->z_norm = (int)val; - if ((error = zfs_get_zplprop(os, ZFS_PROP_UTF8ONLY, &zval)) != 0) - goto out; - zfsvfs->z_utf8 = (zval != 0); + error = zfs_get_zplprop(os, ZFS_PROP_UTF8ONLY, &val); + if (error != 0) + return (error); + zfsvfs->z_utf8 = (val != 0); - if ((error = zfs_get_zplprop(os, ZFS_PROP_CASE, &zval)) != 0) - goto out; - zfsvfs->z_case = (uint_t)zval; + error = zfs_get_zplprop(os, ZFS_PROP_CASE, &val); + if (error != 0) + return (error); + zfsvfs->z_case = (uint_t)val; /* * Fold case on file systems that are always or sometimes case @@ -925,60 +899,99 @@ zfsvfs_create(const char *osname, zfsvfs_t **zfvp) zfsvfs->z_use_fuids = USE_FUIDS(zfsvfs->z_version, zfsvfs->z_os); zfsvfs->z_use_sa = USE_SA(zfsvfs->z_version, zfsvfs->z_os); + uint64_t sa_obj = 0; if (zfsvfs->z_use_sa) { /* should either have both of these objects or none */ error = zap_lookup(os, MASTER_NODE_OBJ, ZFS_SA_ATTRS, 8, 1, &sa_obj); - if (error) - goto out; - } else { - /* - * Pre SA versions file systems should never touch - * either the attribute registration or layout objects. - */ - sa_obj = 0; + if (error != 0) + return (error); } error = sa_setup(os, sa_obj, zfs_attr_table, ZPL_END, &zfsvfs->z_attr_table); - if (error) - goto out; + if (error != 0) + return (error); if (zfsvfs->z_version >= ZPL_VERSION_SA) sa_register_update_callback(os, zfs_sa_upgrade); error = zap_lookup(os, MASTER_NODE_OBJ, ZFS_ROOT_OBJ, 8, 1, &zfsvfs->z_root); - if (error) - goto out; + if (error != 0) + return (error); ASSERT(zfsvfs->z_root != 0); error = zap_lookup(os, MASTER_NODE_OBJ, ZFS_UNLINKED_SET, 8, 1, &zfsvfs->z_unlinkedobj); - if (error) - goto out; + if (error != 0) + return (error); error = zap_lookup(os, MASTER_NODE_OBJ, zfs_userquota_prop_prefixes[ZFS_PROP_USERQUOTA], 8, 1, &zfsvfs->z_userquota_obj); - if (error && error != ENOENT) - goto out; + if (error == ENOENT) + zfsvfs->z_userquota_obj = 0; + else if (error != 0) + return (error); error = zap_lookup(os, MASTER_NODE_OBJ, zfs_userquota_prop_prefixes[ZFS_PROP_GROUPQUOTA], 8, 1, &zfsvfs->z_groupquota_obj); - if (error && error != ENOENT) - goto out; + if (error == ENOENT) + zfsvfs->z_groupquota_obj = 0; + else if (error != 0) + return (error); error = zap_lookup(os, MASTER_NODE_OBJ, ZFS_FUID_TABLES, 8, 1, &zfsvfs->z_fuid_obj); - if (error && error != ENOENT) - goto out; + if (error == ENOENT) + zfsvfs->z_fuid_obj = 0; + else if (error != 0) + return (error); error = zap_lookup(os, MASTER_NODE_OBJ, ZFS_SHARES_DIR, 8, 1, &zfsvfs->z_shares_dir); - if (error && error != ENOENT) - goto out; + if (error == ENOENT) + zfsvfs->z_shares_dir = 0; + else if (error != 0) + return (error); + + return (0); +} + +int +zfsvfs_create(const char *osname, zfsvfs_t **zfvp) +{ + objset_t *os; + zfsvfs_t *zfsvfs; + int error; + + /* + * XXX: Fix struct statfs so this isn't necessary! + * + * The 'osname' is used as the filesystem's special node, which means + * it must fit in statfs.f_mntfromname, or else it can't be + * enumerated, so libzfs_mnttab_find() returns NULL, which causes + * 'zfs unmount' to think it's not mounted when it is. + */ + if (strlen(osname) >= MNAMELEN) + return (SET_ERROR(ENAMETOOLONG)); + + zfsvfs = kmem_zalloc(sizeof (zfsvfs_t), KM_SLEEP); + + /* + * We claim to always be readonly so we can open snapshots; + * other ZPL code will prevent us from writing to snapshots. + */ + error = dmu_objset_own(osname, DMU_OST_ZFS, B_TRUE, zfsvfs, &os); + if (error) { + kmem_free(zfsvfs, sizeof (zfsvfs_t)); + return (error); + } + + zfsvfs->z_vfs = NULL; + zfsvfs->z_parent = zfsvfs; mutex_init(&zfsvfs->z_znodes_lock, NULL, MUTEX_DEFAULT, NULL); mutex_init(&zfsvfs->z_lock, NULL, MUTEX_DEFAULT, NULL); @@ -987,17 +1000,19 @@ zfsvfs_create(const char *osname, zfsvfs_t **zfvp) rrm_init(&zfsvfs->z_teardown_lock, B_FALSE); rw_init(&zfsvfs->z_teardown_inactive_lock, NULL, RW_DEFAULT, NULL); rw_init(&zfsvfs->z_fuid_lock, NULL, RW_DEFAULT, NULL); - for (i = 0; i != ZFS_OBJ_MTX_SZ; i++) + for (int i = 0; i != ZFS_OBJ_MTX_SZ; i++) mutex_init(&zfsvfs->z_hold_mtx[i], NULL, MUTEX_DEFAULT, NULL); + error = zfsvfs_init(zfsvfs, os); + if (error != 0) { + dmu_objset_disown(os, zfsvfs); + *zfvp = NULL; + kmem_free(zfsvfs, sizeof (zfsvfs_t)); + return (error); + } + *zfvp = zfsvfs; return (0); - -out: - dmu_objset_disown(os, zfsvfs); - *zfvp = NULL; - kmem_free(zfsvfs, sizeof (zfsvfs_t)); - return (error); } static int @@ -2199,7 +2214,6 @@ zfs_resume_fs(zfsvfs_t *zfsvfs, const char *osname) { int err; znode_t *zp; - uint64_t sa_obj = 0; ASSERT(RRM_WRITE_HELD(&zfsvfs->z_teardown_lock)); ASSERT(RW_WRITE_HELD(&zfsvfs->z_teardown_inactive_lock)); @@ -2208,35 +2222,16 @@ zfs_resume_fs(zfsvfs_t *zfsvfs, const char *osname) * We already own this, so just hold and rele it to update the * objset_t, as the one we had before may have been evicted. */ - VERIFY0(dmu_objset_hold(osname, zfsvfs, &zfsvfs->z_os)); - VERIFY3P(zfsvfs->z_os->os_dsl_dataset->ds_owner, ==, zfsvfs); - VERIFY(dsl_dataset_long_held(zfsvfs->z_os->os_dsl_dataset)); - dmu_objset_rele(zfsvfs->z_os, zfsvfs); - - /* - * Make sure version hasn't changed - */ - - err = zfs_get_zplprop(zfsvfs->z_os, ZFS_PROP_VERSION, - &zfsvfs->z_version); - - if (err) - goto bail; - - err = zap_lookup(zfsvfs->z_os, MASTER_NODE_OBJ, - ZFS_SA_ATTRS, 8, 1, &sa_obj); - - if (err && zfsvfs->z_version >= ZPL_VERSION_SA) - goto bail; + objset_t *os; + VERIFY0(dmu_objset_hold(osname, zfsvfs, &os)); + VERIFY3P(os->os_dsl_dataset->ds_owner, ==, zfsvfs); + VERIFY(dsl_dataset_long_held(os->os_dsl_dataset)); + dmu_objset_rele(os, zfsvfs); - if ((err = sa_setup(zfsvfs->z_os, sa_obj, - zfs_attr_table, ZPL_END, &zfsvfs->z_attr_table)) != 0) + err = zfsvfs_init(zfsvfs, os); + if (err != 0) goto bail; - if (zfsvfs->z_version >= ZPL_VERSION_SA) - sa_register_update_callback(zfsvfs->z_os, - zfs_sa_upgrade); - VERIFY(zfsvfs_setup(zfsvfs, B_FALSE) == 0); zfs_set_fuid_feature(zfsvfs); diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_vnops.c b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_vnops.c index b0f11ac..17179f6 100644 --- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_vnops.c +++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_vnops.c @@ -2438,6 +2438,7 @@ top: dmu_tx_hold_zap(tx, zfsvfs->z_unlinkedobj, FALSE, NULL); zfs_sa_upgrade_txholds(tx, zp); zfs_sa_upgrade_txholds(tx, dzp); + dmu_tx_mark_netfree(tx); error = dmu_tx_assign(tx, waited ? TXG_WAITED : TXG_NOWAIT); if (error) { rw_exit(&zp->z_parent_lock); diff --git a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zio.c b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zio.c index 616aa79..98fd449 100644 --- a/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zio.c +++ b/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zio.c @@ -20,7 +20,7 @@ */ /* * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved. - * Copyright (c) 2011, 2015 by Delphix. All rights reserved. + * Copyright (c) 2011, 2016 by Delphix. All rights reserved. * Copyright (c) 2011 Nexenta Systems, Inc. All rights reserved. * Copyright (c) 2014 Integros [integros.com] */ @@ -771,9 +771,10 @@ zio_read(zio_t *pio, spa_t *spa, const blkptr_t *bp, zio_t * zio_write(zio_t *pio, spa_t *spa, uint64_t txg, blkptr_t *bp, void *data, uint64_t size, const zio_prop_t *zp, - zio_done_func_t *ready, zio_done_func_t *physdone, zio_done_func_t *done, - void *private, - zio_priority_t priority, enum zio_flag flags, const zbookmark_phys_t *zb) + zio_done_func_t *ready, zio_done_func_t *children_ready, + zio_done_func_t *physdone, zio_done_func_t *done, + void *private, zio_priority_t priority, enum zio_flag flags, + const zbookmark_phys_t *zb) { zio_t *zio; @@ -792,6 +793,7 @@ zio_write(zio_t *pio, spa_t *spa, uint64_t txg, blkptr_t *bp, ZIO_DDT_CHILD_WRITE_PIPELINE : ZIO_WRITE_PIPELINE); zio->io_ready = ready; + zio->io_children_ready = children_ready; zio->io_physdone = physdone; zio->io_prop = *zp; @@ -1189,6 +1191,16 @@ zio_write_bp_init(zio_t *zio) if (!IO_IS_ALLOCATING(zio)) return (ZIO_PIPELINE_CONTINUE); + if (zio->io_children_ready != NULL) { + /* + * Now that all our children are ready, run the callback + * associated with this zio in case it wants to modify the + * data to be written. + */ + ASSERT3U(zp->zp_level, >, 0); + zio->io_children_ready(zio); + } + ASSERT(zio->io_child_type != ZIO_CHILD_DDT); if (zio->io_bp_override) { @@ -2118,9 +2130,9 @@ zio_write_gang_block(zio_t *pio) zio_nowait(zio_write(zio, spa, txg, &gbh->zg_blkptr[g], (char *)pio->io_data + (pio->io_size - resid), lsize, &zp, - zio_write_gang_member_ready, NULL, NULL, &gn->gn_child[g], - pio->io_priority, ZIO_GANG_CHILD_FLAGS(pio), - &pio->io_bookmark)); + zio_write_gang_member_ready, NULL, NULL, NULL, + &gn->gn_child[g], pio->io_priority, + ZIO_GANG_CHILD_FLAGS(pio), &pio->io_bookmark)); } /* @@ -2509,7 +2521,7 @@ zio_ddt_write(zio_t *zio) dio = zio_write(zio, spa, txg, bp, zio->io_orig_data, zio->io_orig_size, &czp, NULL, NULL, - zio_ddt_ditto_write_done, dde, zio->io_priority, + NULL, zio_ddt_ditto_write_done, dde, zio->io_priority, ZIO_DDT_CHILD_FLAGS(zio), &zio->io_bookmark); zio_push_transform(dio, zio->io_data, zio->io_size, 0, NULL); @@ -2530,7 +2542,8 @@ zio_ddt_write(zio_t *zio) ddt_phys_addref(ddp); } else { cio = zio_write(zio, spa, txg, bp, zio->io_orig_data, - zio->io_orig_size, zp, zio_ddt_child_write_ready, NULL, + zio->io_orig_size, zp, + zio_ddt_child_write_ready, NULL, NULL, zio_ddt_child_write_done, dde, zio->io_priority, ZIO_DDT_CHILD_FLAGS(zio), &zio->io_bookmark); diff --git a/sys/dev/mlx5/mlx5_en/mlx5_en_main.c b/sys/dev/mlx5/mlx5_en/mlx5_en_main.c index 4a8029b..03d9c15 100644 --- a/sys/dev/mlx5/mlx5_en/mlx5_en_main.c +++ b/sys/dev/mlx5/mlx5_en/mlx5_en_main.c @@ -3114,6 +3114,13 @@ mlx5e_destroy_ifp(struct mlx5_core_dev *mdev, void *vpriv) /* don't allow more IOCTLs */ priv->gone = 1; + /* + * Clear the device description to avoid use after free, + * because the bsddev is not destroyed when this module is + * unloaded: + */ + device_set_desc(mdev->pdev->dev.bsddev, NULL); + /* XXX wait a bit to allow IOCTL handlers to complete */ pause("W", hz); diff --git a/sys/dev/usb/input/ukbd.c b/sys/dev/usb/input/ukbd.c index 3ee6b8bf..506ee45 100644 --- a/sys/dev/usb/input/ukbd.c +++ b/sys/dev/usb/input/ukbd.c @@ -201,6 +201,7 @@ struct ukbd_softc { int sc_mode; /* input mode (K_XLATE,K_RAW,K_CODE) */ int sc_state; /* shift/lock key state */ int sc_accents; /* accent key index (> 0) */ + int sc_polling; /* polling recursion count */ int sc_led_size; int sc_kbd_size; @@ -1986,7 +1987,16 @@ ukbd_poll(keyboard_t *kbd, int on) struct ukbd_softc *sc = kbd->kb_data; UKBD_LOCK(); - if (on) { + /* + * Keep a reference count on polling to allow recursive + * cngrab() during a panic for example. + */ + if (on) + sc->sc_polling++; + else + sc->sc_polling--; + + if (sc->sc_polling != 0) { sc->sc_flags |= UKBD_FLAG_POLLING; sc->sc_poll_thread = curthread; } else { diff --git a/sys/dev/usb/usbdevs b/sys/dev/usb/usbdevs index a53e15b..84db769 100644 --- a/sys/dev/usb/usbdevs +++ b/sys/dev/usb/usbdevs @@ -2726,6 +2726,7 @@ product LOGITECH BB13 0xc401 USB-PS/2 Trackball product LOGITECH RK53 0xc501 Cordless mouse product LOGITECH RB6 0xc503 Cordless keyboard product LOGITECH MX700 0xc506 Cordless optical mouse +product LOGITECH UNIFYING 0xc52b Logitech Unifying Receiver product LOGITECH QUICKCAMPRO2 0xd001 QuickCam Pro /* Logitec Corp. products */ diff --git a/sys/dev/virtio/network/if_vtnet.c b/sys/dev/virtio/network/if_vtnet.c index a5e6383..8474edb 100644 --- a/sys/dev/virtio/network/if_vtnet.c +++ b/sys/dev/virtio/network/if_vtnet.c @@ -228,18 +228,32 @@ static void vtnet_disable_interrupts(struct vtnet_softc *); static int vtnet_tunable_int(struct vtnet_softc *, const char *, int); /* Tunables. */ +static SYSCTL_NODE(_hw, OID_AUTO, vtnet, CTLFLAG_RD, 0, "VNET driver parameters"); static int vtnet_csum_disable = 0; TUNABLE_INT("hw.vtnet.csum_disable", &vtnet_csum_disable); +SYSCTL_INT(_hw_vtnet, OID_AUTO, csum_disable, CTLFLAG_RDTUN, + &vtnet_csum_disable, 0, "Disables receive and send checksum offload"); static int vtnet_tso_disable = 0; TUNABLE_INT("hw.vtnet.tso_disable", &vtnet_tso_disable); +SYSCTL_INT(_hw_vtnet, OID_AUTO, tso_disable, CTLFLAG_RDTUN, &vtnet_tso_disable, + 0, "Disables TCP Segmentation Offload"); static int vtnet_lro_disable = 0; TUNABLE_INT("hw.vtnet.lro_disable", &vtnet_lro_disable); +SYSCTL_INT(_hw_vtnet, OID_AUTO, lro_disable, CTLFLAG_RDTUN, &vtnet_lro_disable, + 0, "Disables TCP Large Receive Offload"); static int vtnet_mq_disable = 0; TUNABLE_INT("hw.vtnet.mq_disable", &vtnet_mq_disable); -static int vtnet_mq_max_pairs = 0; +SYSCTL_INT(_hw_vtnet, OID_AUTO, mq_disable, CTLFLAG_RDTUN, &vtnet_mq_disable, + 0, "Disables Multi Queue support"); +static int vtnet_mq_max_pairs = VTNET_MAX_QUEUE_PAIRS; TUNABLE_INT("hw.vtnet.mq_max_pairs", &vtnet_mq_max_pairs); +SYSCTL_INT(_hw_vtnet, OID_AUTO, mq_max_pairs, CTLFLAG_RDTUN, + &vtnet_mq_max_pairs, 0, "Sets the maximum number of Multi Queue pairs"); static int vtnet_rx_process_limit = 512; TUNABLE_INT("hw.vtnet.rx_process_limit", &vtnet_rx_process_limit); +SYSCTL_INT(_hw_vtnet, OID_AUTO, rx_process_limit, CTLFLAG_RDTUN, + &vtnet_rx_process_limit, 0, + "Limits the number RX segments processed in a single pass"); static uma_zone_t vtnet_tx_header_zone; @@ -589,7 +603,6 @@ static void vtnet_setup_features(struct vtnet_softc *sc) { device_t dev; - int max_pairs, max; dev = sc->vtnet_dev; @@ -638,32 +651,31 @@ vtnet_setup_features(struct vtnet_softc *sc) if (virtio_with_feature(dev, VIRTIO_NET_F_MQ) && sc->vtnet_flags & VTNET_FLAG_CTRL_VQ) { - max_pairs = virtio_read_dev_config_2(dev, + sc->vtnet_max_vq_pairs = virtio_read_dev_config_2(dev, offsetof(struct virtio_net_config, max_virtqueue_pairs)); - if (max_pairs < VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MIN || - max_pairs > VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MAX) - max_pairs = 1; } else - max_pairs = 1; + sc->vtnet_max_vq_pairs = 1; - if (max_pairs > 1) { + if (sc->vtnet_max_vq_pairs > 1) { /* - * Limit the maximum number of queue pairs to the number of - * CPUs or the configured maximum. The actual number of - * queues that get used may be less. + * Limit the maximum number of queue pairs to the lower of + * the number of CPUs and the configured maximum. + * The actual number of queues that get used may be less. */ + int max; + max = vtnet_tunable_int(sc, "mq_max_pairs", vtnet_mq_max_pairs); - if (max > 0 && max_pairs > max) - max_pairs = max; - if (max_pairs > mp_ncpus) - max_pairs = mp_ncpus; - if (max_pairs > VTNET_MAX_QUEUE_PAIRS) - max_pairs = VTNET_MAX_QUEUE_PAIRS; - if (max_pairs > 1) - sc->vtnet_flags |= VTNET_FLAG_MULTIQ; + if (max > VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MIN) { + if (max > mp_ncpus) + max = mp_ncpus; + if (max > VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MAX) + max = VIRTIO_NET_CTRL_MQ_VQ_PAIRS_MAX; + if (max > 1) { + sc->vtnet_requested_vq_pairs = max; + sc->vtnet_flags |= VTNET_FLAG_MULTIQ; + } + } } - - sc->vtnet_max_vq_pairs = max_pairs; } static int @@ -2989,13 +3001,11 @@ vtnet_set_active_vq_pairs(struct vtnet_softc *sc) dev = sc->vtnet_dev; if ((sc->vtnet_flags & VTNET_FLAG_MULTIQ) == 0) { - MPASS(sc->vtnet_max_vq_pairs == 1); sc->vtnet_act_vq_pairs = 1; return; } - /* BMV: Just use the maximum configured for now. */ - npairs = sc->vtnet_max_vq_pairs; + npairs = sc->vtnet_requested_vq_pairs; if (vtnet_ctrl_mq_cmd(sc, npairs) != 0) { device_printf(dev, @@ -3851,6 +3861,9 @@ vtnet_setup_sysctl(struct vtnet_softc *sc) SYSCTL_ADD_INT(ctx, child, OID_AUTO, "max_vq_pairs", CTLFLAG_RD, &sc->vtnet_max_vq_pairs, 0, "Maximum number of supported virtqueue pairs"); + SYSCTL_ADD_INT(ctx, child, OID_AUTO, "requested_vq_pairs", + CTLFLAG_RD, &sc->vtnet_requested_vq_pairs, 0, + "Requested number of virtqueue pairs"); SYSCTL_ADD_INT(ctx, child, OID_AUTO, "act_vq_pairs", CTLFLAG_RD, &sc->vtnet_act_vq_pairs, 0, "Number of active virtqueue pairs"); diff --git a/sys/dev/virtio/network/if_vtnetvar.h b/sys/dev/virtio/network/if_vtnetvar.h index f89f6b1..15436d9 100644 --- a/sys/dev/virtio/network/if_vtnetvar.h +++ b/sys/dev/virtio/network/if_vtnetvar.h @@ -155,6 +155,7 @@ struct vtnet_softc { int vtnet_if_flags; int vtnet_act_vq_pairs; int vtnet_max_vq_pairs; + int vtnet_requested_vq_pairs; struct virtqueue *vtnet_ctrl_vq; struct vtnet_mac_filter *vtnet_mac_filter; diff --git a/sys/fs/tmpfs/tmpfs_subr.c b/sys/fs/tmpfs/tmpfs_subr.c index f06958e..9781b2c 100644 --- a/sys/fs/tmpfs/tmpfs_subr.c +++ b/sys/fs/tmpfs/tmpfs_subr.c @@ -818,10 +818,13 @@ tmpfs_dir_lookup_cookie(struct tmpfs_node *node, off_t cookie, goto out; } - MPASS((cookie & TMPFS_DIRCOOKIE_MASK) == cookie); - dekey.td_hash = cookie; - /* Recover if direntry for cookie was removed */ - de = RB_NFIND(tmpfs_dir, dirhead, &dekey); + if ((cookie & TMPFS_DIRCOOKIE_MASK) != cookie) { + de = NULL; + } else { + dekey.td_hash = cookie; + /* Recover if direntry for cookie was removed */ + de = RB_NFIND(tmpfs_dir, dirhead, &dekey); + } dc->tdc_tree = de; dc->tdc_current = de; if (de != NULL && tmpfs_dirent_duphead(de)) { diff --git a/sys/i386/i386/trap.c b/sys/i386/i386/trap.c index 2fcf690..a186ff5 100644 --- a/sys/i386/i386/trap.c +++ b/sys/i386/i386/trap.c @@ -534,8 +534,8 @@ trap(struct trapframe *frame) case T_DNA: #ifdef DEV_NPX - KASSERT(!PCB_USER_FPU(td->td_pcb), - ("Unregistered use of FPU in kernel")); + if (PCB_USER_FPU(td->td_pcb)) + panic("Unregistered use of FPU in kernel"); if (npxdna()) goto out; #endif diff --git a/sys/i386/linux/linux_ptrace.c b/sys/i386/linux/linux_ptrace.c index 2925e6b..fbe66b4 100644 --- a/sys/i386/linux/linux_ptrace.c +++ b/sys/i386/linux/linux_ptrace.c @@ -69,7 +69,7 @@ __FBSDID("$FreeBSD$"); #define PTRACE_ATTACH 16 #define PTRACE_DETACH 17 -#define PTRACE_SYSCALL 24 +#define LINUX_PTRACE_SYSCALL 24 #define PTRACE_GETREGS 12 #define PTRACE_SETREGS 13 @@ -473,7 +473,7 @@ linux_ptrace(struct thread *td, struct linux_ptrace_args *uap) break; } - case PTRACE_SYSCALL: + case LINUX_PTRACE_SYSCALL: /* fall through */ default: printf("linux: ptrace(%u, ...) not implemented\n", diff --git a/sys/kern/kern_exec.c b/sys/kern/kern_exec.c index 8357ab8..6b34066 100644 --- a/sys/kern/kern_exec.c +++ b/sys/kern/kern_exec.c @@ -55,6 +55,7 @@ __FBSDID("$FreeBSD$"); #include <sys/priv.h> #include <sys/proc.h> #include <sys/pioctl.h> +#include <sys/ptrace.h> #include <sys/namei.h> #include <sys/resourcevar.h> #include <sys/rwlock.h> @@ -899,7 +900,8 @@ exec_fail_dealloc: if (error == 0) { PROC_LOCK(p); - td->td_dbgflags |= TDB_EXEC; + if (p->p_ptevents & PTRACE_EXEC) + td->td_dbgflags |= TDB_EXEC; PROC_UNLOCK(p); /* diff --git a/sys/kern/kern_exit.c b/sys/kern/kern_exit.c index f9244af..c362530 100644 --- a/sys/kern/kern_exit.c +++ b/sys/kern/kern_exit.c @@ -336,6 +336,7 @@ exit1(struct thread *td, int rv) rv = p->p_xstat; /* Event handler could change exit status */ stopprofclock(p); p->p_flag &= ~(P_TRACED | P_PPWAIT | P_PPTRACE); + p->p_ptevents = 0; /* * Stop the real interval timer. If the handler is currently @@ -519,8 +520,12 @@ exit1(struct thread *td, int rv) */ clear_orphan(q); q->p_flag &= ~(P_TRACED | P_STOPPED_TRACE); - FOREACH_THREAD_IN_PROC(q, tdt) - tdt->td_dbgflags &= ~TDB_SUSPEND; + q->p_flag2 &= ~P2_PTRACE_FSTP; + q->p_ptevents = 0; + FOREACH_THREAD_IN_PROC(q, tdt) { + tdt->td_dbgflags &= ~(TDB_SUSPEND | TDB_XSIG | + TDB_FSTP); + } kern_psignal(q, SIGKILL); } PROC_UNLOCK(q); diff --git a/sys/kern/kern_fork.c b/sys/kern/kern_fork.c index a84b619..4d96840 100644 --- a/sys/kern/kern_fork.c +++ b/sys/kern/kern_fork.c @@ -411,6 +411,7 @@ do_fork(struct thread *td, int flags, struct proc *p2, struct thread *td2, __rangeof(struct proc, p_startzero, p_endzero)); p2->p_treeflag = 0; p2->p_filemon = NULL; + p2->p_ptevents = 0; /* Tell the prison that we exist. */ prison_proc_hold(p2->p_ucred->cr_prison); @@ -710,8 +711,7 @@ do_fork(struct thread *td, int flags, struct proc *p2, struct thread *td2, if ((flags & RFMEM) == 0 && dtrace_fasttrap_fork) dtrace_fasttrap_fork(p1, p2); #endif - if ((p1->p_flag & (P_TRACED | P_FOLLOWFORK)) == (P_TRACED | - P_FOLLOWFORK)) { + if (p1->p_ptevents & PTRACE_FORK) { /* * Arrange for debugger to receive the fork event. * @@ -1057,21 +1057,19 @@ fork_return(struct thread *td, struct trapframe *frame) if (td->td_dbgflags & TDB_STOPATFORK) { sx_xlock(&proctree_lock); PROC_LOCK(p); - if ((p->p_pptr->p_flag & (P_TRACED | P_FOLLOWFORK)) == - (P_TRACED | P_FOLLOWFORK)) { + if (p->p_pptr->p_ptevents & PTRACE_FORK) { /* * If debugger still wants auto-attach for the * parent's children, do it now. */ dbg = p->p_pptr->p_pptr; - p->p_flag |= P_TRACED; - p->p_oppid = p->p_pptr->p_pid; + proc_set_traced(p); CTR2(KTR_PTRACE, "fork_return: attaching to new child pid %d: oppid %d", p->p_pid, p->p_oppid); proc_reparent(p, dbg); sx_xunlock(&proctree_lock); - td->td_dbgflags |= TDB_CHILD | TDB_SCX; + td->td_dbgflags |= TDB_CHILD | TDB_SCX | TDB_FSTP; ptracestop(td, SIGSTOP); td->td_dbgflags &= ~(TDB_CHILD | TDB_SCX); } else { @@ -1083,7 +1081,7 @@ fork_return(struct thread *td, struct trapframe *frame) cv_broadcast(&p->p_dbgwait); } PROC_UNLOCK(p); - } else if (p->p_flag & P_TRACED) { + } else if (p->p_flag & P_TRACED || td->td_dbgflags & TDB_BORN) { /* * This is the start of a new thread in a traced * process. Report a system call exit event. @@ -1091,9 +1089,10 @@ fork_return(struct thread *td, struct trapframe *frame) PROC_LOCK(p); td->td_dbgflags |= TDB_SCX; _STOPEVENT(p, S_SCX, td->td_dbg_sc_code); - if ((p->p_stops & S_PT_SCX) != 0) + if ((p->p_ptevents & PTRACE_SCX) != 0 || + (td->td_dbgflags & TDB_BORN) != 0) ptracestop(td, SIGTRAP); - td->td_dbgflags &= ~TDB_SCX; + td->td_dbgflags &= ~(TDB_SCX | TDB_BORN); PROC_UNLOCK(p); } diff --git a/sys/kern/kern_kthread.c b/sys/kern/kern_kthread.c index 70d95fa..fb46025 100644 --- a/sys/kern/kern_kthread.c +++ b/sys/kern/kern_kthread.c @@ -322,11 +322,13 @@ void kthread_exit(void) { struct proc *p; + struct thread *td; - p = curthread->td_proc; + td = curthread; + p = td->td_proc; /* A module may be waiting for us to exit. */ - wakeup(curthread); + wakeup(td); /* * The last exiting thread in a kernel process must tear down @@ -339,9 +341,10 @@ kthread_exit(void) rw_wunlock(&tidhash_lock); kproc_exit(0); } - LIST_REMOVE(curthread, td_hash); + LIST_REMOVE(td, td_hash); rw_wunlock(&tidhash_lock); - umtx_thread_exit(curthread); + umtx_thread_exit(td); + tdsigcleanup(td); PROC_SLOCK(p); thread_exit(); } diff --git a/sys/kern/kern_sig.c b/sys/kern/kern_sig.c index 29783f8..2c37d76 100644 --- a/sys/kern/kern_sig.c +++ b/sys/kern/kern_sig.c @@ -2175,9 +2175,10 @@ tdsendsignal(struct proc *p, struct thread *td, int sig, ksiginfo_t *ksi) !((prop & SA_CONT) && (p->p_flag & P_STOPPED_SIG))) return (ret); /* - * SIGKILL: Remove procfs STOPEVENTs. + * SIGKILL: Remove procfs STOPEVENTs and ptrace events. */ if (sig == SIGKILL) { + p->p_ptevents = 0; /* from procfs_ioctl.c: PIOCBIC */ p->p_stops = 0; /* from procfs_ioctl.c: PIOCCONT */ @@ -2488,19 +2489,36 @@ ptracestop(struct thread *td, int sig) td->td_tid, p->p_pid, td->td_dbgflags, sig); PROC_SLOCK(p); while ((p->p_flag & P_TRACED) && (td->td_dbgflags & TDB_XSIG)) { - if (p->p_flag & P_SINGLE_EXIT) { + if (p->p_flag & P_SINGLE_EXIT && + !(td->td_dbgflags & TDB_EXIT)) { + /* + * Ignore ptrace stops except for thread exit + * events when the process exits. + */ td->td_dbgflags &= ~TDB_XSIG; PROC_SUNLOCK(p); return (sig); } + /* - * Just make wait() to work, the last stopped thread - * will win. + * Make wait(2) work. Ensure that right after the + * attach, the thread which was decided to become the + * leader of attach gets reported to the waiter. + * Otherwise, just avoid overwriting another thread's + * assignment to p_xthread. If another thread has + * already set p_xthread, the current thread will get + * a chance to report itself upon the next iteration. */ - p->p_xstat = sig; - p->p_xthread = td; - p->p_flag |= (P_STOPPED_SIG|P_STOPPED_TRACE); - sig_suspend_threads(td, p, 0); + if ((td->td_dbgflags & TDB_FSTP) != 0 || + ((p->p_flag & P2_PTRACE_FSTP) == 0 && + p->p_xthread == NULL)) { + p->p_xstat = sig; + p->p_xthread = td; + td->td_dbgflags &= ~TDB_FSTP; + p->p_flag2 &= ~P2_PTRACE_FSTP; + p->p_flag |= P_STOPPED_SIG | P_STOPPED_TRACE; + sig_suspend_threads(td, p, 0); + } if ((td->td_dbgflags & TDB_STOPATFORK) != 0) { td->td_dbgflags &= ~TDB_STOPATFORK; cv_broadcast(&p->p_dbgwait); @@ -2651,7 +2669,20 @@ issignal(struct thread *td) SIG_STOPSIGMASK(sigpending); if (SIGISEMPTY(sigpending)) /* no signal to send */ return (0); - sig = sig_ffs(&sigpending); + if ((p->p_flag & (P_TRACED | P_PPTRACE)) == P_TRACED && + (p->p_flag2 & P2_PTRACE_FSTP) != 0 && + SIGISMEMBER(sigpending, SIGSTOP)) { + /* + * If debugger just attached, always consume + * SIGSTOP from ptrace(PT_ATTACH) first, to + * execute the debugger attach ritual in + * order. + */ + sig = SIGSTOP; + td->td_dbgflags |= TDB_FSTP; + } else { + sig = sig_ffs(&sigpending); + } if (p->p_stops & S_SIG) { mtx_unlock(&ps->ps_mtx); @@ -2668,7 +2699,7 @@ issignal(struct thread *td) sigqueue_delete(&p->p_sigqueue, sig); continue; } - if (p->p_flag & P_TRACED && (p->p_flag & P_PPTRACE) == 0) { + if ((p->p_flag & (P_TRACED | P_PPTRACE)) == P_TRACED) { /* * If traced, always stop. * Remove old signal from queue before the stop. @@ -2766,6 +2797,8 @@ issignal(struct thread *td) mtx_unlock(&ps->ps_mtx); WITNESS_WARN(WARN_GIANTOK | WARN_SLEEPOK, &p->p_mtx.lock_object, "Catching SIGSTOP"); + sigqueue_delete(&td->td_sigqueue, sig); + sigqueue_delete(&p->p_sigqueue, sig); p->p_flag |= P_STOPPED_SIG; p->p_xstat = sig; PROC_SLOCK(p); @@ -2773,7 +2806,7 @@ issignal(struct thread *td) thread_suspend_switch(td, p); PROC_SUNLOCK(p); mtx_lock(&ps->ps_mtx); - break; + goto next; } else if (prop & SA_IGNORE) { /* * Except for SIGCONT, shouldn't get here. @@ -2804,6 +2837,7 @@ issignal(struct thread *td) } sigqueue_delete(&td->td_sigqueue, sig); /* take the signal! */ sigqueue_delete(&p->p_sigqueue, sig); +next:; } /* NOTREACHED */ } diff --git a/sys/kern/kern_thr.c b/sys/kern/kern_thr.c index 98965b1..b01aecb 100644 --- a/sys/kern/kern_thr.c +++ b/sys/kern/kern_thr.c @@ -36,6 +36,7 @@ __FBSDID("$FreeBSD$"); #include <sys/priv.h> #include <sys/proc.h> #include <sys/posix4.h> +#include <sys/ptrace.h> #include <sys/racct.h> #include <sys/resourcevar.h> #include <sys/rwlock.h> @@ -252,6 +253,8 @@ thread_create(struct thread *td, struct rtprio *rtp, thread_unlock(td); if (P_SHOULDSTOP(p)) newtd->td_flags |= TDF_ASTPENDING | TDF_NEEDSUSPCHK; + if (p->p_ptevents & PTRACE_LWP) + newtd->td_dbgflags |= TDB_BORN; PROC_UNLOCK(p); tidhash_add(newtd); @@ -314,29 +317,54 @@ kern_thr_exit(struct thread *td) p = td->td_proc; - rw_wlock(&tidhash_lock); + /* + * If all of the threads in a process call this routine to + * exit (e.g. all threads call pthread_exit()), exactly one + * thread should return to the caller to terminate the process + * instead of the thread. + * + * Checking p_numthreads alone is not sufficient since threads + * might be committed to terminating while the PROC_LOCK is + * dropped in either ptracestop() or while removing this thread + * from the tidhash. Instead, the p_pendingexits field holds + * the count of threads in either of those states and a thread + * is considered the "last" thread if all of the other threads + * in a process are already terminating. + */ PROC_LOCK(p); - - if (p->p_numthreads != 1) { - racct_sub(p, RACCT_NTHR, 1); - LIST_REMOVE(td, td_hash); - rw_wunlock(&tidhash_lock); - tdsigcleanup(td); - umtx_thread_exit(td); - PROC_SLOCK(p); - thread_stopped(p); - thread_exit(); - /* NOTREACHED */ + if (p->p_numthreads == p->p_pendingexits + 1) { + /* + * Ignore attempts to shut down last thread in the + * proc. This will actually call _exit(2) in the + * usermode trampoline when it returns. + */ + PROC_UNLOCK(p); + return (0); } + p->p_pendingexits++; + td->td_dbgflags |= TDB_EXIT; + if (p->p_ptevents & PTRACE_LWP) + ptracestop(td, SIGTRAP); + PROC_UNLOCK(p); + tidhash_remove(td); + PROC_LOCK(p); + p->p_pendingexits--; + /* - * Ignore attempts to shut down last thread in the proc. This - * will actually call _exit(2) in the usermode trampoline when - * it returns. + * The check above should prevent all other threads from this + * process from exiting while the PROC_LOCK is dropped, so + * there must be at least one other thread other than the + * current thread. */ - PROC_UNLOCK(p); - rw_wunlock(&tidhash_lock); - return (0); + KASSERT(p->p_numthreads > 1, ("too few threads")); + racct_sub(p, RACCT_NTHR, 1); + tdsigcleanup(td); + umtx_thread_exit(td); + PROC_SLOCK(p); + thread_stopped(p); + thread_exit(); + /* NOTREACHED */ } int diff --git a/sys/kern/subr_syscall.c b/sys/kern/subr_syscall.c index 1ad7dd0..f2b83f0 100644 --- a/sys/kern/subr_syscall.c +++ b/sys/kern/subr_syscall.c @@ -88,7 +88,7 @@ syscallenter(struct thread *td, struct syscall_args *sa) PROC_LOCK(p); td->td_dbg_sc_code = sa->code; td->td_dbg_sc_narg = sa->narg; - if (p->p_stops & S_PT_SCE) + if (p->p_ptevents & PTRACE_SCE) ptracestop((td), SIGTRAP); PROC_UNLOCK(p); } @@ -215,7 +215,7 @@ syscallret(struct thread *td, int error, struct syscall_args *sa __unused) */ if (traced && ((td->td_dbgflags & (TDB_FORK | TDB_EXEC)) != 0 || - (p->p_stops & S_PT_SCX) != 0)) + (p->p_ptevents & PTRACE_SCX) != 0)) ptracestop(td, SIGTRAP); td->td_dbgflags &= ~(TDB_SCX | TDB_EXEC | TDB_FORK); PROC_UNLOCK(p); diff --git a/sys/kern/sys_process.c b/sys/kern/sys_process.c index 96e0181..c4533ce 100644 --- a/sys/kern/sys_process.c +++ b/sys/kern/sys_process.c @@ -542,6 +542,7 @@ sys_ptrace(struct thread *td, struct ptrace_args *uap) struct ptrace_lwpinfo32 pl32; struct ptrace_vm_entry32 pve32; #endif + int ptevents; } r; void *addr; int error = 0; @@ -556,6 +557,7 @@ sys_ptrace(struct thread *td, struct ptrace_args *uap) AUDIT_ARG_VALUE(uap->data); addr = &r; switch (uap->req) { + case PT_GET_EVENT_MASK: case PT_GETREGS: case PT_GETFPREGS: case PT_GETDBREGS: @@ -570,6 +572,12 @@ sys_ptrace(struct thread *td, struct ptrace_args *uap) case PT_SETDBREGS: error = COPYIN(uap->addr, &r.dbreg, sizeof r.dbreg); break; + case PT_SET_EVENT_MASK: + if (uap->data != sizeof(r.ptevents)) + error = EINVAL; + else + error = copyin(uap->addr, &r.ptevents, uap->data); + break; case PT_IO: error = COPYIN(uap->addr, &r.piod, sizeof r.piod); break; @@ -603,7 +611,12 @@ sys_ptrace(struct thread *td, struct ptrace_args *uap) case PT_GETDBREGS: error = COPYOUT(&r.dbreg, uap->addr, sizeof r.dbreg); break; + case PT_GET_EVENT_MASK: + /* NB: The size in uap->data is validated in kern_ptrace(). */ + error = copyout(&r.ptevents, uap->addr, uap->data); + break; case PT_LWPINFO: + /* NB: The size in uap->data is validated in kern_ptrace(). */ error = copyout(&r.pl, uap->addr, uap->data); break; } @@ -635,6 +648,17 @@ sys_ptrace(struct thread *td, struct ptrace_args *uap) #define PROC_WRITE(w, t, a) proc_write_ ## w (t, a) #endif +void +proc_set_traced(struct proc *p) +{ + + PROC_LOCK_ASSERT(p, MA_OWNED); + p->p_flag |= P_TRACED; + p->p_flag2 |= P2_PTRACE_FSTP; + p->p_ptevents = PTRACE_DEFAULT; + p->p_oppid = p->p_pptr->p_pid; +} + int kern_ptrace(struct thread *td, int req, pid_t pid, void *addr, int data) { @@ -666,6 +690,9 @@ kern_ptrace(struct thread *td, int req, pid_t pid, void *addr, int data) case PT_TO_SCX: case PT_SYSCALL: case PT_FOLLOW_FORK: + case PT_LWP_EVENTS: + case PT_GET_EVENT_MASK: + case PT_SET_EVENT_MASK: case PT_DETACH: sx_xlock(&proctree_lock); proctree_locked = 1; @@ -840,10 +867,9 @@ kern_ptrace(struct thread *td, int req, pid_t pid, void *addr, int data) switch (req) { case PT_TRACE_ME: /* set my trace flag and "owner" so it can read/write me */ - p->p_flag |= P_TRACED; + proc_set_traced(p); if (p->p_flag & P_PPWAIT) p->p_flag |= P_PPTRACE; - p->p_oppid = p->p_pptr->p_pid; CTR1(KTR_PTRACE, "PT_TRACE_ME: pid %d", p->p_pid); break; @@ -858,8 +884,7 @@ kern_ptrace(struct thread *td, int req, pid_t pid, void *addr, int data) * The old parent is remembered so we can put things back * on a "detach". */ - p->p_flag |= P_TRACED; - p->p_oppid = p->p_pptr->p_pid; + proc_set_traced(p); if (p->p_pptr != td->td_proc) { proc_reparent(p, td->td_proc); } @@ -897,14 +922,50 @@ kern_ptrace(struct thread *td, int req, pid_t pid, void *addr, int data) case PT_FOLLOW_FORK: CTR3(KTR_PTRACE, "PT_FOLLOW_FORK: pid %d %s -> %s", p->p_pid, - p->p_flag & P_FOLLOWFORK ? "enabled" : "disabled", + p->p_ptevents & PTRACE_FORK ? "enabled" : "disabled", + data ? "enabled" : "disabled"); + if (data) + p->p_ptevents |= PTRACE_FORK; + else + p->p_ptevents &= ~PTRACE_FORK; + break; + + case PT_LWP_EVENTS: + CTR3(KTR_PTRACE, "PT_LWP_EVENTS: pid %d %s -> %s", p->p_pid, + p->p_ptevents & PTRACE_LWP ? "enabled" : "disabled", data ? "enabled" : "disabled"); if (data) - p->p_flag |= P_FOLLOWFORK; + p->p_ptevents |= PTRACE_LWP; else - p->p_flag &= ~P_FOLLOWFORK; + p->p_ptevents &= ~PTRACE_LWP; break; + case PT_GET_EVENT_MASK: + if (data != sizeof(p->p_ptevents)) { + error = EINVAL; + break; + } + CTR2(KTR_PTRACE, "PT_GET_EVENT_MASK: pid %d mask %#x", p->p_pid, + p->p_ptevents); + *(int *)addr = p->p_ptevents; + break; + + case PT_SET_EVENT_MASK: + if (data != sizeof(p->p_ptevents)) { + error = EINVAL; + break; + } + tmp = *(int *)addr; + if ((tmp & ~(PTRACE_EXEC | PTRACE_SCE | PTRACE_SCX | + PTRACE_FORK | PTRACE_LWP)) != 0) { + error = EINVAL; + break; + } + CTR3(KTR_PTRACE, "PT_SET_EVENT_MASK: pid %d mask %#x -> %#x", + p->p_pid, p->p_ptevents, tmp); + p->p_ptevents = tmp; + break; + case PT_STEP: case PT_CONTINUE: case PT_TO_SCE: @@ -937,24 +998,24 @@ kern_ptrace(struct thread *td, int req, pid_t pid, void *addr, int data) } switch (req) { case PT_TO_SCE: - p->p_stops |= S_PT_SCE; + p->p_ptevents |= PTRACE_SCE; CTR4(KTR_PTRACE, - "PT_TO_SCE: pid %d, stops = %#x, PC = %#lx, sig = %d", - p->p_pid, p->p_stops, + "PT_TO_SCE: pid %d, events = %#x, PC = %#lx, sig = %d", + p->p_pid, p->p_ptevents, (u_long)(uintfptr_t)addr, data); break; case PT_TO_SCX: - p->p_stops |= S_PT_SCX; + p->p_ptevents |= PTRACE_SCX; CTR4(KTR_PTRACE, - "PT_TO_SCX: pid %d, stops = %#x, PC = %#lx, sig = %d", - p->p_pid, p->p_stops, + "PT_TO_SCX: pid %d, events = %#x, PC = %#lx, sig = %d", + p->p_pid, p->p_ptevents, (u_long)(uintfptr_t)addr, data); break; case PT_SYSCALL: - p->p_stops |= S_PT_SCE | S_PT_SCX; + p->p_ptevents |= PTRACE_SYSCALL; CTR4(KTR_PTRACE, - "PT_SYSCALL: pid %d, stops = %#x, PC = %#lx, sig = %d", - p->p_pid, p->p_stops, + "PT_SYSCALL: pid %d, events = %#x, PC = %#lx, sig = %d", + p->p_pid, p->p_ptevents, (u_long)(uintfptr_t)addr, data); break; case PT_CONTINUE: @@ -973,7 +1034,7 @@ kern_ptrace(struct thread *td, int req, pid_t pid, void *addr, int data) * parent. Otherwise the debugee will be set * as an orphan of the debugger. */ - p->p_flag &= ~(P_TRACED | P_WAITED | P_FOLLOWFORK); + p->p_flag &= ~(P_TRACED | P_WAITED); if (p->p_oppid != p->p_pptr->p_pid) { PROC_LOCK(p->p_pptr); sigqueue_take(p->p_ksi); @@ -990,7 +1051,18 @@ kern_ptrace(struct thread *td, int req, pid_t pid, void *addr, int data) CTR2(KTR_PTRACE, "PT_DETACH: pid %d, sig %d", p->p_pid, data); p->p_oppid = 0; - p->p_stops = 0; + p->p_ptevents = 0; + FOREACH_THREAD_IN_PROC(p, td3) { + if ((td3->td_dbgflags & TDB_FSTP) != 0) { + sigqueue_delete(&td3->td_sigqueue, + SIGSTOP); + } + td3->td_dbgflags &= ~(TDB_XSIG | TDB_FSTP); + } + if ((p->p_flag2 & P2_PTRACE_FSTP) != 0) { + sigqueue_delete(&p->p_sigqueue, SIGSTOP); + p->p_flag2 &= ~P2_PTRACE_FSTP; + } /* should we send SIGCHLD? */ /* childproc_continued(p); */ @@ -1011,7 +1083,7 @@ kern_ptrace(struct thread *td, int req, pid_t pid, void *addr, int data) if (req == PT_DETACH) { FOREACH_THREAD_IN_PROC(p, td3) - td3->td_dbgflags &= ~TDB_SUSPEND; + td3->td_dbgflags &= ~TDB_SUSPEND; } /* * unsuspend all threads, to not let a thread run, @@ -1227,6 +1299,10 @@ kern_ptrace(struct thread *td, int req, pid_t pid, void *addr, int data) } if (td2->td_dbgflags & TDB_CHILD) pl->pl_flags |= PL_FLAG_CHILD; + if (td2->td_dbgflags & TDB_BORN) + pl->pl_flags |= PL_FLAG_BORN; + if (td2->td_dbgflags & TDB_EXIT) + pl->pl_flags |= PL_FLAG_EXITED; pl->pl_sigmask = td2->td_sigmask; pl->pl_siglist = td2->td_siglist; strcpy(pl->pl_tdname, td2->td_name); diff --git a/sys/netpfil/pf/pf_ioctl.c b/sys/netpfil/pf/pf_ioctl.c index b00952c..2543879 100644 --- a/sys/netpfil/pf/pf_ioctl.c +++ b/sys/netpfil/pf/pf_ioctl.c @@ -3626,7 +3626,9 @@ pf_check_in(void *arg, struct mbuf **m, struct ifnet *ifp, int dir, *m = NULL; } - return (chk); + if (chk != PF_PASS) + return (EACCES); + return (0); } static int @@ -3641,7 +3643,9 @@ pf_check_out(void *arg, struct mbuf **m, struct ifnet *ifp, int dir, *m = NULL; } - return (chk); + if (chk != PF_PASS) + return (EACCES); + return (0); } #endif @@ -3664,7 +3668,9 @@ pf_check6_in(void *arg, struct mbuf **m, struct ifnet *ifp, int dir, m_freem(*m); *m = NULL; } - return chk; + if (chk != PF_PASS) + return (EACCES); + return (0); } static int @@ -3680,7 +3686,9 @@ pf_check6_out(void *arg, struct mbuf **m, struct ifnet *ifp, int dir, m_freem(*m); *m = NULL; } - return chk; + if (chk != PF_PASS) + return (EACCES); + return (0); } #endif /* INET6 */ diff --git a/sys/sys/proc.h b/sys/sys/proc.h index b2de02d..1b8bda5 100644 --- a/sys/sys/proc.h +++ b/sys/sys/proc.h @@ -396,6 +396,9 @@ do { \ #define TDB_STOPATFORK 0x00000080 /* Stop at the return from fork (child only) */ #define TDB_CHILD 0x00000100 /* New child indicator for ptrace() */ +#define TDB_BORN 0x00000200 /* New LWP indicator for ptrace() */ +#define TDB_EXIT 0x00000400 /* Exiting LWP indicator for ptrace() */ +#define TDB_FSTP 0x00001000 /* The thread is PT_ATTACH leader */ /* * "Private" flags kept in td_pflags: @@ -551,6 +554,7 @@ struct proc { int p_pendingcnt; /* how many signals are pending */ struct itimers *p_itimers; /* (c) POSIX interval timers. */ struct procdesc *p_procdesc; /* (e) Process descriptor, if any. */ + int p_pendingexits; /* (c) Count of pending thread exits. */ /* End area that is zeroed on creation. */ #define p_endzero p_magic @@ -608,6 +612,7 @@ struct proc { our subtree. */ struct pgrp *p_pgrp; /* (c + e) Pointer to process group. */ struct filemon *p_filemon; /* (c) filemon-specific data. */ + u_int p_ptevents; /* (c) ptrace() event mask. */ }; #define p_session p_pgrp->pg_session @@ -635,7 +640,7 @@ struct proc { #define P_ADVLOCK 0x00001 /* Process may hold a POSIX advisory lock. */ #define P_CONTROLT 0x00002 /* Has a controlling terminal. */ #define P_KTHREAD 0x00004 /* Kernel thread (*). */ -#define P_FOLLOWFORK 0x00008 /* Attach parent debugger to children. */ +#define P_UNUSED3 0x00008 /* --available-- */ #define P_PPWAIT 0x00010 /* Parent is waiting for child to exec/exit. */ #define P_PROFIL 0x00020 /* Has started profiling. */ #define P_STOPPROF 0x00040 /* Has thread requesting to stop profiling. */ @@ -674,6 +679,7 @@ struct proc { #define P2_NOTRACE 0x00000002 /* No ptrace(2) attach or coredumps. */ #define P2_NOTRACE_EXEC 0x00000004 /* Keep P2_NOPTRACE on exec(2). */ #define P2_AST_SU 0x00000008 /* Handles SU ast for kthreads. */ +#define P2_PTRACE_FSTP 0x00000010 /* SIGSTOP from PT_ATTACH not yet handled. */ /* Flags protected by proctree_lock, kept in p_treeflags. */ #define P_TREE_ORPHANED 0x00000001 /* Reparented, on orphan list */ @@ -926,6 +932,7 @@ void proc_linkup(struct proc *p, struct thread *td); struct proc *proc_realparent(struct proc *child); void proc_reap(struct thread *td, struct proc *p, int *status, int options); void proc_reparent(struct proc *child, struct proc *newparent); +void proc_set_traced(struct proc *p); struct pstats *pstats_alloc(void); void pstats_fork(struct pstats *src, struct pstats *dst); void pstats_free(struct pstats *ps); diff --git a/sys/sys/ptrace.h b/sys/sys/ptrace.h index 7135767..e2b6a5f 100644 --- a/sys/sys/ptrace.h +++ b/sys/sys/ptrace.h @@ -64,6 +64,10 @@ #define PT_SYSCALL 22 #define PT_FOLLOW_FORK 23 +#define PT_LWP_EVENTS 24 /* report LWP birth and exit */ + +#define PT_GET_EVENT_MASK 25 /* get mask of optional events */ +#define PT_SET_EVENT_MASK 26 /* set mask of optional events */ #define PT_GETREGS 33 /* get general-purpose registers */ #define PT_SETREGS 34 /* set general-purpose registers */ @@ -78,6 +82,16 @@ #define PT_FIRSTMACH 64 /* for machine-specific requests */ #include <machine/ptrace.h> /* machine-specific requests, if any */ +/* Events used with PT_GET_EVENT_MASK and PT_SET_EVENT_MASK */ +#define PTRACE_EXEC 0x0001 +#define PTRACE_SCE 0x0002 +#define PTRACE_SCX 0x0004 +#define PTRACE_SYSCALL (PTRACE_SCE | PTRACE_SCX) +#define PTRACE_FORK 0x0008 +#define PTRACE_LWP 0x0010 + +#define PTRACE_DEFAULT (PTRACE_EXEC) + struct ptrace_io_desc { int piod_op; /* I/O operation */ void *piod_offs; /* child offset */ @@ -108,6 +122,8 @@ struct ptrace_lwpinfo { #define PL_FLAG_SI 0x20 /* siginfo is valid */ #define PL_FLAG_FORKED 0x40 /* new child */ #define PL_FLAG_CHILD 0x80 /* I am from child */ +#define PL_FLAG_BORN 0x100 /* new LWP */ +#define PL_FLAG_EXITED 0x200 /* exiting LWP */ sigset_t pl_sigmask; /* LWP signal mask */ sigset_t pl_siglist; /* LWP pending signal */ struct __siginfo pl_siginfo; /* siginfo for signal */ @@ -133,13 +149,6 @@ struct ptrace_vm_entry { #ifdef _KERNEL -/* - * The flags below are used for ptrace(2) tracing and have no relation - * to procfs. They are stored in struct proc's p_stops member. - */ -#define S_PT_SCE 0x000010000 -#define S_PT_SCX 0x000020000 - int ptrace_set_pc(struct thread *_td, unsigned long _addr); int ptrace_single_step(struct thread *_td); int ptrace_clear_single_step(struct thread *_td); diff --git a/sys/vm/uma_core.c b/sys/vm/uma_core.c index ee0b207..11dceeb 100644 --- a/sys/vm/uma_core.c +++ b/sys/vm/uma_core.c @@ -2770,6 +2770,8 @@ zfree_start: goto zfree_start; } cache->uc_freebucket = NULL; + /* We are no longer associated with this CPU. */ + critical_exit(); /* Can we throw this on the zone full list? */ if (bucket != NULL) { @@ -2782,9 +2784,6 @@ zfree_start: LIST_INSERT_HEAD(&zone->uz_buckets, bucket, ub_link); } - /* We are no longer associated with this CPU. */ - critical_exit(); - /* * We bump the uz count when the cache size is insufficient to * handle the working set. diff --git a/sys/vm/vm_page.c b/sys/vm/vm_page.c index 7010580..433c875 100644 --- a/sys/vm/vm_page.c +++ b/sys/vm/vm_page.c @@ -2651,7 +2651,7 @@ vm_page_advise(vm_page_t m, int advice) * But we do make the page as freeable as we can without * actually taking the step of unmapping it. */ - m->dirty = 0; + vm_page_undirty(m); else if (advice != MADV_DONTNEED) return; diff --git a/sys/vm/vm_pageout.c b/sys/vm/vm_pageout.c index 0d81be4..c9c1271 100644 --- a/sys/vm/vm_pageout.c +++ b/sys/vm/vm_pageout.c @@ -262,7 +262,7 @@ vm_pageout_init_marker(vm_page_t marker, u_short queue) * * Lock vm object currently associated with `m'. VM_OBJECT_TRYWLOCK is * known to have failed and page queue must be either PQ_ACTIVE or - * PQ_INACTIVE. To avoid lock order violation, unlock the page queues + * PQ_INACTIVE. To avoid lock order violation, unlock the page queue * while locking the vm object. Use marker page to detect page queue * changes and maintain notion of next page on page queue. Return * TRUE if no changes were detected, FALSE otherwise. vm object is @@ -950,7 +950,7 @@ vm_pageout_scan(struct vm_domain *vmd, int pass) int vnodes_skipped = 0; int maxlaunder, scan_tick, scanned, starting_page_shortage; int lockmode; - boolean_t queues_locked; + boolean_t queue_locked; /* * If we need to reclaim memory ask kernel caches to return @@ -1015,12 +1015,12 @@ vm_pageout_scan(struct vm_domain *vmd, int pass) pq = &vmd->vmd_pagequeues[PQ_INACTIVE]; maxscan = pq->pq_cnt; vm_pagequeue_lock(pq); - queues_locked = TRUE; + queue_locked = TRUE; for (m = TAILQ_FIRST(&pq->pq_pl); m != NULL && maxscan-- > 0 && page_shortage > 0; m = next) { vm_pagequeue_assert_locked(pq); - KASSERT(queues_locked, ("unlocked queues")); + KASSERT(queue_locked, ("unlocked inactive queue")); KASSERT(m->queue == PQ_INACTIVE, ("Inactive queue %p", m)); PCPU_INC(cnt.v_pdpages); @@ -1076,7 +1076,7 @@ vm_pageout_scan(struct vm_domain *vmd, int pass) */ TAILQ_INSERT_AFTER(&pq->pq_pl, m, &vmd->vmd_marker, plinks.q); vm_pagequeue_unlock(pq); - queues_locked = FALSE; + queue_locked = FALSE; /* * We bump the activation count if the page has been @@ -1109,12 +1109,12 @@ vm_pageout_scan(struct vm_domain *vmd, int pass) m->act_count += act_delta + ACT_ADVANCE; } else { vm_pagequeue_lock(pq); - queues_locked = TRUE; + queue_locked = TRUE; vm_page_requeue_locked(m); } VM_OBJECT_WUNLOCK(object); vm_page_unlock(m); - goto relock_queues; + goto relock_queue; } if (m->hold_count != 0) { @@ -1129,7 +1129,7 @@ vm_pageout_scan(struct vm_domain *vmd, int pass) * loop over the active queue below. */ addl_page_shortage++; - goto relock_queues; + goto relock_queue; } /* @@ -1175,7 +1175,7 @@ vm_pageout_scan(struct vm_domain *vmd, int pass) */ m->flags |= PG_WINATCFLS; vm_pagequeue_lock(pq); - queues_locked = TRUE; + queue_locked = TRUE; vm_page_requeue_locked(m); } else if (maxlaunder > 0) { /* @@ -1206,9 +1206,9 @@ vm_pageout_scan(struct vm_domain *vmd, int pass) vm_pagequeue_lock(pq); vm_page_unlock(m); VM_OBJECT_WUNLOCK(object); - queues_locked = TRUE; + queue_locked = TRUE; vm_page_requeue_locked(m); - goto relock_queues; + goto relock_queue; } /* @@ -1263,7 +1263,7 @@ vm_pageout_scan(struct vm_domain *vmd, int pass) VM_OBJECT_WLOCK(object); vm_page_lock(m); vm_pagequeue_lock(pq); - queues_locked = TRUE; + queue_locked = TRUE; /* * The page might have been moved to another * queue during potential blocking in vget() @@ -1303,7 +1303,7 @@ vm_pageout_scan(struct vm_domain *vmd, int pass) goto unlock_and_continue; } vm_pagequeue_unlock(pq); - queues_locked = FALSE; + queue_locked = FALSE; } /* @@ -1324,9 +1324,9 @@ unlock_and_continue: vm_page_lock_assert(m, MA_NOTOWNED); VM_OBJECT_WUNLOCK(object); if (mp != NULL) { - if (queues_locked) { + if (queue_locked) { vm_pagequeue_unlock(pq); - queues_locked = FALSE; + queue_locked = FALSE; } if (vp != NULL) vput(vp); @@ -1334,14 +1334,14 @@ unlock_and_continue: vn_finished_write(mp); } vm_page_lock_assert(m, MA_NOTOWNED); - goto relock_queues; + goto relock_queue; } vm_page_unlock(m); VM_OBJECT_WUNLOCK(object); -relock_queues: - if (!queues_locked) { +relock_queue: + if (!queue_locked) { vm_pagequeue_lock(pq); - queues_locked = TRUE; + queue_locked = TRUE; } next = TAILQ_NEXT(&vmd->vmd_marker, plinks.q); TAILQ_REMOVE(&pq->pq_pl, &vmd->vmd_marker, plinks.q); diff --git a/tests/sys/kern/ptrace_test.c b/tests/sys/kern/ptrace_test.c index 1be1144..44ab15b 100644 --- a/tests/sys/kern/ptrace_test.c +++ b/tests/sys/kern/ptrace_test.c @@ -1008,6 +1008,16 @@ simple_thread(void *arg __unused) pthread_exit(NULL); } +static __dead2 void +simple_thread_main(void) +{ + pthread_t thread; + + CHILD_REQUIRE(pthread_create(&thread, NULL, simple_thread, NULL) == 0); + CHILD_REQUIRE(pthread_join(thread, NULL) == 0); + exit(1); +} + /* * Verify that pl_syscall_code in struct ptrace_lwpinfo for a new * thread reports the correct value. @@ -1022,14 +1032,8 @@ ATF_TC_BODY(ptrace__new_child_pl_syscall_code_thread, tc) ATF_REQUIRE((fpid = fork()) != -1); if (fpid == 0) { - pthread_t thread; - trace_me(); - - CHILD_REQUIRE(pthread_create(&thread, NULL, simple_thread, - NULL) == 0); - CHILD_REQUIRE(pthread_join(thread, NULL) == 0); - exit(1); + simple_thread_main(); } /* The first wait() should report the stop from SIGSTOP. */ @@ -1092,6 +1096,373 @@ ATF_TC_BODY(ptrace__new_child_pl_syscall_code_thread, tc) ATF_REQUIRE(errno == ECHILD); } +/* + * Verify that the expected LWP events are reported for a child thread. + */ +ATF_TC_WITHOUT_HEAD(ptrace__lwp_events); +ATF_TC_BODY(ptrace__lwp_events, tc) +{ + struct ptrace_lwpinfo pl; + pid_t fpid, wpid; + lwpid_t lwps[2]; + int status; + + ATF_REQUIRE((fpid = fork()) != -1); + if (fpid == 0) { + trace_me(); + simple_thread_main(); + } + + /* The first wait() should report the stop from SIGSTOP. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(wpid == fpid); + ATF_REQUIRE(WIFSTOPPED(status)); + ATF_REQUIRE(WSTOPSIG(status) == SIGSTOP); + + ATF_REQUIRE(ptrace(PT_LWPINFO, wpid, (caddr_t)&pl, + sizeof(pl)) != -1); + lwps[0] = pl.pl_lwpid; + + ATF_REQUIRE(ptrace(PT_LWP_EVENTS, wpid, NULL, 1) == 0); + + /* Continue the child ignoring the SIGSTOP. */ + ATF_REQUIRE(ptrace(PT_CONTINUE, fpid, (caddr_t)1, 0) == 0); + + /* The first event should be for the child thread's birth. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(wpid == fpid); + ATF_REQUIRE(WIFSTOPPED(status)); + ATF_REQUIRE(WSTOPSIG(status) == SIGTRAP); + + ATF_REQUIRE(ptrace(PT_LWPINFO, wpid, (caddr_t)&pl, sizeof(pl)) != -1); + ATF_REQUIRE((pl.pl_flags & (PL_FLAG_BORN | PL_FLAG_SCX)) == + (PL_FLAG_BORN | PL_FLAG_SCX)); + ATF_REQUIRE(pl.pl_lwpid != lwps[0]); + lwps[1] = pl.pl_lwpid; + + ATF_REQUIRE(ptrace(PT_CONTINUE, fpid, (caddr_t)1, 0) == 0); + + /* The next event should be for the child thread's death. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(wpid == fpid); + ATF_REQUIRE(WIFSTOPPED(status)); + ATF_REQUIRE(WSTOPSIG(status) == SIGTRAP); + + ATF_REQUIRE(ptrace(PT_LWPINFO, wpid, (caddr_t)&pl, sizeof(pl)) != -1); + ATF_REQUIRE((pl.pl_flags & (PL_FLAG_EXITED | PL_FLAG_SCE)) == + (PL_FLAG_EXITED | PL_FLAG_SCE)); + ATF_REQUIRE(pl.pl_lwpid == lwps[1]); + + ATF_REQUIRE(ptrace(PT_CONTINUE, fpid, (caddr_t)1, 0) == 0); + + /* The last event should be for the child process's exit. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(WIFEXITED(status)); + ATF_REQUIRE(WEXITSTATUS(status) == 1); + + wpid = wait(&status); + ATF_REQUIRE(wpid == -1); + ATF_REQUIRE(errno == ECHILD); +} + +static void * +exec_thread(void *arg __unused) +{ + + execl("/usr/bin/true", "true", NULL); + exit(127); +} + +static __dead2 void +exec_thread_main(void) +{ + pthread_t thread; + + CHILD_REQUIRE(pthread_create(&thread, NULL, exec_thread, NULL) == 0); + for (;;) + sleep(60); + exit(1); +} + +/* + * Verify that the expected LWP events are reported for a multithreaded + * process that calls execve(2). + */ +ATF_TC_WITHOUT_HEAD(ptrace__lwp_events_exec); +ATF_TC_BODY(ptrace__lwp_events_exec, tc) +{ + struct ptrace_lwpinfo pl; + pid_t fpid, wpid; + lwpid_t lwps[2]; + int status; + + ATF_REQUIRE((fpid = fork()) != -1); + if (fpid == 0) { + trace_me(); + exec_thread_main(); + } + + /* The first wait() should report the stop from SIGSTOP. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(wpid == fpid); + ATF_REQUIRE(WIFSTOPPED(status)); + ATF_REQUIRE(WSTOPSIG(status) == SIGSTOP); + + ATF_REQUIRE(ptrace(PT_LWPINFO, wpid, (caddr_t)&pl, + sizeof(pl)) != -1); + lwps[0] = pl.pl_lwpid; + + ATF_REQUIRE(ptrace(PT_LWP_EVENTS, wpid, NULL, 1) == 0); + + /* Continue the child ignoring the SIGSTOP. */ + ATF_REQUIRE(ptrace(PT_CONTINUE, fpid, (caddr_t)1, 0) == 0); + + /* The first event should be for the child thread's birth. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(wpid == fpid); + ATF_REQUIRE(WIFSTOPPED(status)); + ATF_REQUIRE(WSTOPSIG(status) == SIGTRAP); + + ATF_REQUIRE(ptrace(PT_LWPINFO, wpid, (caddr_t)&pl, sizeof(pl)) != -1); + ATF_REQUIRE((pl.pl_flags & (PL_FLAG_BORN | PL_FLAG_SCX)) == + (PL_FLAG_BORN | PL_FLAG_SCX)); + ATF_REQUIRE(pl.pl_lwpid != lwps[0]); + lwps[1] = pl.pl_lwpid; + + ATF_REQUIRE(ptrace(PT_CONTINUE, fpid, (caddr_t)1, 0) == 0); + + /* + * The next event should be for the main thread's death due to + * single threading from execve(). + */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(wpid == fpid); + ATF_REQUIRE(WIFSTOPPED(status)); + ATF_REQUIRE(WSTOPSIG(status) == SIGTRAP); + + ATF_REQUIRE(ptrace(PT_LWPINFO, wpid, (caddr_t)&pl, sizeof(pl)) != -1); + ATF_REQUIRE((pl.pl_flags & (PL_FLAG_EXITED | PL_FLAG_SCE)) == + (PL_FLAG_EXITED)); + ATF_REQUIRE(pl.pl_lwpid == lwps[0]); + + ATF_REQUIRE(ptrace(PT_CONTINUE, fpid, (caddr_t)1, 0) == 0); + + /* The next event should be for the child process's exec. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(WIFSTOPPED(status)); + ATF_REQUIRE(WSTOPSIG(status) == SIGTRAP); + + ATF_REQUIRE(ptrace(PT_LWPINFO, wpid, (caddr_t)&pl, sizeof(pl)) != -1); + ATF_REQUIRE((pl.pl_flags & (PL_FLAG_EXEC | PL_FLAG_SCX)) == + (PL_FLAG_EXEC | PL_FLAG_SCX)); + ATF_REQUIRE(pl.pl_lwpid == lwps[1]); + + ATF_REQUIRE(ptrace(PT_CONTINUE, fpid, (caddr_t)1, 0) == 0); + + /* The last event should be for the child process's exit. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(WIFEXITED(status)); + ATF_REQUIRE(WEXITSTATUS(status) == 0); + + wpid = wait(&status); + ATF_REQUIRE(wpid == -1); + ATF_REQUIRE(errno == ECHILD); +} + +static void +handler(int sig __unused) +{ +} + +static void +signal_main(void) +{ + + signal(SIGINFO, handler); + raise(SIGINFO); + exit(0); +} + +/* + * Verify that the expected ptrace event is reported for a signal. + */ +ATF_TC_WITHOUT_HEAD(ptrace__siginfo); +ATF_TC_BODY(ptrace__siginfo, tc) +{ + struct ptrace_lwpinfo pl; + pid_t fpid, wpid; + int status; + + ATF_REQUIRE((fpid = fork()) != -1); + if (fpid == 0) { + trace_me(); + signal_main(); + } + + /* The first wait() should report the stop from SIGSTOP. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(wpid == fpid); + ATF_REQUIRE(WIFSTOPPED(status)); + ATF_REQUIRE(WSTOPSIG(status) == SIGSTOP); + + ATF_REQUIRE(ptrace(PT_CONTINUE, fpid, (caddr_t)1, 0) == 0); + + /* The next event should be for the SIGINFO. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(WIFSTOPPED(status)); + ATF_REQUIRE(WSTOPSIG(status) == SIGINFO); + + ATF_REQUIRE(ptrace(PT_LWPINFO, wpid, (caddr_t)&pl, sizeof(pl)) != -1); + ATF_REQUIRE(pl.pl_event == PL_EVENT_SIGNAL); + ATF_REQUIRE(pl.pl_flags & PL_FLAG_SI); + ATF_REQUIRE(pl.pl_siginfo.si_code == SI_LWP); + ATF_REQUIRE(pl.pl_siginfo.si_pid == wpid); + + ATF_REQUIRE(ptrace(PT_CONTINUE, fpid, (caddr_t)1, 0) == 0); + + /* The last event should be for the child process's exit. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(WIFEXITED(status)); + ATF_REQUIRE(WEXITSTATUS(status) == 0); + + wpid = wait(&status); + ATF_REQUIRE(wpid == -1); + ATF_REQUIRE(errno == ECHILD); +} + +/* + * Verify that the expected ptrace events are reported for PTRACE_EXEC. + */ +ATF_TC_WITHOUT_HEAD(ptrace__ptrace_exec_disable); +ATF_TC_BODY(ptrace__ptrace_exec_disable, tc) +{ + pid_t fpid, wpid; + int events, status; + + ATF_REQUIRE((fpid = fork()) != -1); + if (fpid == 0) { + trace_me(); + exec_thread(NULL); + } + + /* The first wait() should report the stop from SIGSTOP. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(wpid == fpid); + ATF_REQUIRE(WIFSTOPPED(status)); + ATF_REQUIRE(WSTOPSIG(status) == SIGSTOP); + + events = 0; + ATF_REQUIRE(ptrace(PT_SET_EVENT_MASK, fpid, (caddr_t)&events, + sizeof(events)) == 0); + + ATF_REQUIRE(ptrace(PT_CONTINUE, fpid, (caddr_t)1, 0) == 0); + + /* Should get one event at exit. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(WIFEXITED(status)); + ATF_REQUIRE(WEXITSTATUS(status) == 0); + + wpid = wait(&status); + ATF_REQUIRE(wpid == -1); + ATF_REQUIRE(errno == ECHILD); +} + +ATF_TC_WITHOUT_HEAD(ptrace__ptrace_exec_enable); +ATF_TC_BODY(ptrace__ptrace_exec_enable, tc) +{ + struct ptrace_lwpinfo pl; + pid_t fpid, wpid; + int events, status; + + ATF_REQUIRE((fpid = fork()) != -1); + if (fpid == 0) { + trace_me(); + exec_thread(NULL); + } + + /* The first wait() should report the stop from SIGSTOP. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(wpid == fpid); + ATF_REQUIRE(WIFSTOPPED(status)); + ATF_REQUIRE(WSTOPSIG(status) == SIGSTOP); + + events = PTRACE_EXEC; + ATF_REQUIRE(ptrace(PT_SET_EVENT_MASK, fpid, (caddr_t)&events, + sizeof(events)) == 0); + + ATF_REQUIRE(ptrace(PT_CONTINUE, fpid, (caddr_t)1, 0) == 0); + + /* The next event should be for the child process's exec. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(WIFSTOPPED(status)); + ATF_REQUIRE(WSTOPSIG(status) == SIGTRAP); + + ATF_REQUIRE(ptrace(PT_LWPINFO, wpid, (caddr_t)&pl, sizeof(pl)) != -1); + ATF_REQUIRE((pl.pl_flags & (PL_FLAG_EXEC | PL_FLAG_SCX)) == + (PL_FLAG_EXEC | PL_FLAG_SCX)); + + ATF_REQUIRE(ptrace(PT_CONTINUE, fpid, (caddr_t)1, 0) == 0); + + /* The last event should be for the child process's exit. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(WIFEXITED(status)); + ATF_REQUIRE(WEXITSTATUS(status) == 0); + + wpid = wait(&status); + ATF_REQUIRE(wpid == -1); + ATF_REQUIRE(errno == ECHILD); +} + +ATF_TC_WITHOUT_HEAD(ptrace__event_mask); +ATF_TC_BODY(ptrace__event_mask, tc) +{ + pid_t fpid, wpid; + int events, status; + + ATF_REQUIRE((fpid = fork()) != -1); + if (fpid == 0) { + trace_me(); + exit(0); + } + + /* The first wait() should report the stop from SIGSTOP. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(wpid == fpid); + ATF_REQUIRE(WIFSTOPPED(status)); + ATF_REQUIRE(WSTOPSIG(status) == SIGSTOP); + + /* PT_FOLLOW_FORK should toggle the state of PTRACE_FORK. */ + ATF_REQUIRE(ptrace(PT_FOLLOW_FORK, fpid, NULL, 1) != -1); + ATF_REQUIRE(ptrace(PT_GET_EVENT_MASK, fpid, (caddr_t)&events, + sizeof(events)) == 0); + ATF_REQUIRE(events & PTRACE_FORK); + ATF_REQUIRE(ptrace(PT_FOLLOW_FORK, fpid, NULL, 0) != -1); + ATF_REQUIRE(ptrace(PT_GET_EVENT_MASK, fpid, (caddr_t)&events, + sizeof(events)) == 0); + ATF_REQUIRE(!(events & PTRACE_FORK)); + + /* PT_LWP_EVENTS should toggle the state of PTRACE_LWP. */ + ATF_REQUIRE(ptrace(PT_LWP_EVENTS, fpid, NULL, 1) != -1); + ATF_REQUIRE(ptrace(PT_GET_EVENT_MASK, fpid, (caddr_t)&events, + sizeof(events)) == 0); + ATF_REQUIRE(events & PTRACE_LWP); + ATF_REQUIRE(ptrace(PT_LWP_EVENTS, fpid, NULL, 0) != -1); + ATF_REQUIRE(ptrace(PT_GET_EVENT_MASK, fpid, (caddr_t)&events, + sizeof(events)) == 0); + ATF_REQUIRE(!(events & PTRACE_LWP)); + + ATF_REQUIRE(ptrace(PT_CONTINUE, fpid, (caddr_t)1, 0) == 0); + + /* Should get one event at exit. */ + wpid = waitpid(fpid, &status, 0); + ATF_REQUIRE(WIFEXITED(status)); + ATF_REQUIRE(WEXITSTATUS(status) == 0); + + wpid = wait(&status); + ATF_REQUIRE(wpid == -1); + ATF_REQUIRE(errno == ECHILD); +} + ATF_TP_ADD_TCS(tp) { @@ -1110,6 +1481,12 @@ ATF_TP_ADD_TCS(tp) ATF_TP_ADD_TC(tp, ptrace__new_child_pl_syscall_code_fork); ATF_TP_ADD_TC(tp, ptrace__new_child_pl_syscall_code_vfork); ATF_TP_ADD_TC(tp, ptrace__new_child_pl_syscall_code_thread); + ATF_TP_ADD_TC(tp, ptrace__lwp_events); + ATF_TP_ADD_TC(tp, ptrace__lwp_events_exec); + ATF_TP_ADD_TC(tp, ptrace__siginfo); + ATF_TP_ADD_TC(tp, ptrace__ptrace_exec_disable); + ATF_TP_ADD_TC(tp, ptrace__ptrace_exec_enable); + ATF_TP_ADD_TC(tp, ptrace__event_mask); return (atf_no_error()); } diff --git a/usr.bin/getconf/getconf.1 b/usr.bin/getconf/getconf.1 index c3baed9..529c666 100644 --- a/usr.bin/getconf/getconf.1 +++ b/usr.bin/getconf/getconf.1 @@ -28,7 +28,7 @@ .\" .\" $FreeBSD$ .\" -.Dd September 18, 2002 +.Dd August 8, 2016 .Dt GETCONF 1 .Os .Sh NAME @@ -122,7 +122,7 @@ Exactly 32-bit integer, long, and pointer; at least 64-bit file offset. .It Li POSIX_V6_LP64_OFF64 Exactly 32-bit integer; exactly 64-bit long, pointer, and file offset. .Sy Supported platforms : -.Tn Alpha , +.Tn AMD64 , .Tn SPARC64 . .It Li POSIX_V6_LPBIG_OFFBIG At least 32-bit integer; at least 64-bit long, pointer, and file offset. diff --git a/usr.bin/getconf/progenv.gperf b/usr.bin/getconf/progenv.gperf index 40ce16a..4810553 100644 --- a/usr.bin/getconf/progenv.gperf +++ b/usr.bin/getconf/progenv.gperf @@ -30,7 +30,7 @@ static const struct map *in_word_set(const char *str); * be updated. (We cheat here and define the supported environments * statically.) */ -#if defined(__alpha__) || defined(__sparc64__) || defined(__amd64__) +#if defined(__sparc64__) || defined(__amd64__) #define have_LP64_OFF64 NULL #endif diff --git a/usr.bin/truss/syscalls.c b/usr.bin/truss/syscalls.c index 985a074..dfbbbac 100644 --- a/usr.bin/truss/syscalls.c +++ b/usr.bin/truss/syscalls.c @@ -283,6 +283,8 @@ static struct syscall decoded_syscalls[] = { .args = { { Atfd, 0 }, { Name, 1 }, { Atfd, 2 }, { Name, 3 } } }, { .name = "rfork", .ret_type = 1, .nargs = 1, .args = { { Rforkflags, 0 } } }, + { .name = "rmdir", .ret_type = 1, .nargs = 1, + .args = { { Name, 0 } } }, { .name = "select", .ret_type = 1, .nargs = 5, .args = { { Int, 0 }, { Fd_set, 1 }, { Fd_set, 2 }, { Fd_set, 3 }, { Timeval, 4 } } }, |