From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by sourceware.org (Postfix) with ESMTPS id 31A253858423 for ; Wed, 5 Jul 2023 15:06:17 +0000 (GMT) DMARC-Filter: OpenDMARC Filter v1.4.2 sourceware.org 31A253858423 Authentication-Results: sourceware.org; dmarc=pass (p=none dis=none) header.from=redhat.com Authentication-Results: sourceware.org; spf=pass smtp.mailfrom=redhat.com DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1688569576; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version:content-type:content-type: in-reply-to:in-reply-to:references:references; bh=uoVaVCQTxsqUY9tuqAGqrL9vyl5wyMWCH9KbN0a2k2w=; b=OHRCiKdhr6+oQRKcWh2S0zPXHyZcuOelsr1mrDX1yteo0O3GEw1EltHeSaDu2ahQGZZkWa Y/kAoQ7oB/U4+VlyPdnSQMneKG2yxIWQmFkvWGM4sOgpe2uVv4mQa8mSPIar5hfU1ke7sR qXpoWQQGV2/LulcY8ADxljCm3Rwx90U= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-649-tPyNV426Pz6SIuzmy0Duvw-1; Wed, 05 Jul 2023 11:06:13 -0400 X-MC-Unique: tPyNV426Pz6SIuzmy0Duvw-1 Received: from smtp.corp.redhat.com (int-mx08.intmail.prod.int.rdu2.redhat.com [10.11.54.8]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 5F305185A7B0; Wed, 5 Jul 2023 15:06:13 +0000 (UTC) Received: from oak (unknown [10.22.8.210]) by smtp.corp.redhat.com (Postfix) with ESMTPS id 3B028C478DE; Wed, 5 Jul 2023 15:06:13 +0000 (UTC) Date: Wed, 5 Jul 2023 11:06:11 -0400 From: Joe Simmons-Talbott To: Adhemerval Zanella Cc: libc-alpha@sourceware.org Subject: Re: [PATCH v3] vfscanf-internal: Remove potentially unbounded allocas Message-ID: <20230705150611.GX6392@oak> References: <20230703194008.2213843-1-adhemerval.zanella@linaro.org> MIME-Version: 1.0 In-Reply-To: <20230703194008.2213843-1-adhemerval.zanella@linaro.org> X-Scanned-By: MIMEDefang 3.1 on 10.11.54.8 X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: redhat.com Content-Type: text/plain; charset=us-ascii Content-Disposition: inline X-Spam-Status: No, score=-11.7 required=5.0 tests=BAYES_00,DKIMWL_WL_HIGH,DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,DKIM_VALID_EF,GIT_PATCH_0,KAM_SHORT,RCVD_IN_DNSWL_NONE,RCVD_IN_MSPIKE_H4,RCVD_IN_MSPIKE_WL,SPF_HELO_NONE,SPF_NONE,TXREP,T_SCC_BODY_TEXT_LINE autolearn=ham autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on server2.sourceware.org List-Id: On Mon, Jul 03, 2023 at 04:40:08PM -0300, Adhemerval Zanella wrote: > Some locales define a list of mapping pairs of alternate digits and > separators for input digits (to_inpunct). This require the scanf > to create a list of all possible inputs for the optional type > modifier 'I'. > > Checked on x86_64-linux-gnu. > --- > Changes from v2: > * Remove ENOMEM set for malloc failure. > > Changes from v1: > * Add malloc failure check. LGTM. Thanks! Reviewed-by: Joe Simmons-Talbott > --- > stdio-common/Makefile | 3 ++ > stdio-common/tst-scanf-to_inpunct.c | 78 ++++++++++++++++++++++++++++ > stdio-common/vfscanf-internal.c | 51 ++++++++++++------- > wcsmbs/Makefile | 3 ++ > wcsmbs/tst-wscanf-to_inpunct.c | 79 +++++++++++++++++++++++++++++ > 5 files changed, 197 insertions(+), 17 deletions(-) > create mode 100644 stdio-common/tst-scanf-to_inpunct.c > create mode 100644 wcsmbs/tst-wscanf-to_inpunct.c > > diff --git a/stdio-common/Makefile b/stdio-common/Makefile > index 8871ec7668..f6d9017ff1 100644 > --- a/stdio-common/Makefile > +++ b/stdio-common/Makefile > @@ -231,6 +231,7 @@ tests := \ > tst-scanf-binary-gnu11 \ > tst-scanf-binary-gnu89 \ > tst-scanf-round \ > + tst-scanf-to_inpunct \ > tst-setvbuf1 \ > tst-sprintf \ > tst-sprintf-errno \ > @@ -347,6 +348,7 @@ LOCALES := \ > de_DE.ISO-8859-1 \ > de_DE.UTF-8 \ > en_US.ISO-8859-1 \ > + fa_IR.UTF-8 \ > hi_IN.UTF-8 \ > ja_JP.EUC-JP \ > ps_AF.UTF-8 \ > @@ -366,6 +368,7 @@ $(objpfx)tst-swprintf.out: $(gen-locales) > $(objpfx)tst-vfprintf-mbs-prec.out: $(gen-locales) > $(objpfx)tst-vfprintf-width-i18n.out: $(gen-locales) > $(objpfx)tst-grouping3.out: $(gen-locales) > +$(objpfx)tst-scanf-to_inpunct.out: $(gen-locales) > endif > > tst-printf-bz18872-ENV = MALLOC_TRACE=$(objpfx)tst-printf-bz18872.mtrace \ > diff --git a/stdio-common/tst-scanf-to_inpunct.c b/stdio-common/tst-scanf-to_inpunct.c > new file mode 100644 > index 0000000000..32236ac2dc > --- /dev/null > +++ b/stdio-common/tst-scanf-to_inpunct.c > @@ -0,0 +1,78 @@ > +/* Test scanf for languages with mapping pairs of alternate digits and > + separators. > + Copyright (C) 2023 Free Software Foundation, Inc. > + This file is part of the GNU C Library. > + > + The GNU C Library is free software; you can redistribute it and/or > + modify it under the terms of the GNU Lesser General Public > + License as published by the Free Software Foundation; either > + version 2.1 of the License, or (at your option) any later version. > + > + The GNU C Library is distributed in the hope that it will be useful, > + but WITHOUT ANY WARRANTY; without even the implied warranty of > + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU > + Lesser General Public License for more details. > + > + You should have received a copy of the GNU Lesser General Public > + License along with the GNU C Library; if not, see > + . */ > + > +#include > +#include > +#include > +#include > + > +/* fa_IR defines to_inpunct for numbers. */ > +static const struct > +{ > + int n; > + const char *str; > +} inputs[] = > +{ > + { 1, "\xdb\xb1" }, > + { 2, "\xdb\xb2" }, > + { 3, "\xdb\xb3" }, > + { 4, "\xdb\xb4" }, > + { 5, "\xdb\xb5" }, > + { 6, "\xdb\xb6" }, > + { 7, "\xdb\xb7" }, > + { 8, "\xdb\xb8" }, > + { 9, "\xdb\xb9" }, > + { 10, "\xdb\xb1\xdb\xb0" }, > + { 11, "\xdb\xb1\xdb\xb1" }, > + { 12, "\xdb\xb1\xdb\xb2" }, > + { 13, "\xdb\xb1\xdb\xb3" }, > + { 14, "\xdb\xb1\xdb\xb4" }, > + { 15, "\xdb\xb1\xdb\xb5" }, > + { 16, "\xdb\xb1\xdb\xb6" }, > + { 17, "\xdb\xb1\xdb\xb7" }, > + { 18, "\xdb\xb1\xdb\xb8" }, > + { 19, "\xdb\xb1\xdb\xb9" }, > + { 20, "\xdb\xb2\xdb\xb0" }, > + { 30, "\xdb\xb3\xdb\xb0" }, > + { 40, "\xdb\xb4\xdb\xb0" }, > + { 50, "\xdb\xb5\xdb\xb0" }, > + { 60, "\xdb\xb6\xdb\xb0" }, > + { 70, "\xdb\xb7\xdb\xb0" }, > + { 80, "\xdb\xb8\xdb\xb0" }, > + { 90, "\xdb\xb9\xdb\xb0" }, > + { 100, "\xdb\xb1\xdb\xb0\xdb\xb0" }, > + { 1000, "\xdb\xb1\xdb\xb0\xdb\xb0\xdb\xb0" }, > +}; > + > +static int > +do_test (void) > +{ > + xsetlocale (LC_ALL, "fa_IR.UTF-8"); > + > + for (int i = 0; i < array_length (inputs); i++) > + { > + int n; > + sscanf (inputs[i].str, "%Id", &n); > + TEST_COMPARE (n, inputs[i].n); > + } > + > + return 0; > +} > + > +#include > diff --git a/stdio-common/vfscanf-internal.c b/stdio-common/vfscanf-internal.c > index bfb9baa21a..9b1197d751 100644 > --- a/stdio-common/vfscanf-internal.c > +++ b/stdio-common/vfscanf-internal.c > @@ -1455,13 +1455,14 @@ __vfscanf_internal (FILE *s, const char *format, va_list argptr, > int from_level; > int to_level; > int level; > + enum { num_digits_len = 10 }; > #ifdef COMPILE_WSCANF > - const wchar_t *wcdigits[10]; > - const wchar_t *wcdigits_extended[10]; > + const wchar_t *wcdigits[num_digits_len]; > #else > - const char *mbdigits[10]; > - const char *mbdigits_extended[10]; > + const char *mbdigits[num_digits_len]; > #endif > + CHAR_T *digits_extended[num_digits_len] = { NULL }; > + > /* "to_inpunct" is a map from ASCII digits to their > equivalent in locale. This is defined for locales > which use an extra digits set. */ > @@ -1482,18 +1483,23 @@ __vfscanf_internal (FILE *s, const char *format, va_list argptr, > /* Adding new level for extra digits set in locale file. */ > ++to_level; > > - for (n = 0; n < 10; ++n) > + for (n = 0; n < num_digits_len; ++n) > { > #ifdef COMPILE_WSCANF > wcdigits[n] = (const wchar_t *) > _NL_CURRENT (LC_CTYPE, _NL_CTYPE_INDIGITS0_WC + n); > > wchar_t *wc_extended = (wchar_t *) > - alloca ((to_level + 2) * sizeof (wchar_t)); > + malloc ((to_level + 2) * sizeof (wchar_t)); > + if (wc_extended == NULL) > + { > + done = EOF; > + goto digits_extended_fail; > + } > __wmemcpy (wc_extended, wcdigits[n], to_level); > wc_extended[to_level] = __towctrans (L'0' + n, map); > wc_extended[to_level + 1] = '\0'; > - wcdigits_extended[n] = wc_extended; > + digits_extended[n] = wc_extended; > #else > mbdigits[n] > = curctype->values[_NL_CTYPE_INDIGITS0_MB + n].string; > @@ -1524,14 +1530,18 @@ __vfscanf_internal (FILE *s, const char *format, va_list argptr, > size_t mbdigits_len = last_char - mbdigits[n]; > > /* Allocate memory for extended multibyte digit. */ > - char *mb_extended; > - mb_extended = (char *) alloca (mbdigits_len + mblen + 1); > + char *mb_extended = malloc (mbdigits_len + mblen + 1); > + if (mb_extended == NULL) > + { > + done = EOF; > + goto digits_extended_fail; > + } > > /* And get the mbdigits + extra_digit string. */ > *(char *) __mempcpy (__mempcpy (mb_extended, mbdigits[n], > mbdigits_len), > extra_mbdigit, mblen) = '\0'; > - mbdigits_extended[n] = mb_extended; > + digits_extended[n] = mb_extended; > #endif > } > } > @@ -1541,7 +1551,7 @@ __vfscanf_internal (FILE *s, const char *format, va_list argptr, > { > /* In this round we get the pointer to the digit strings > and also perform the first round of comparisons. */ > - for (n = 0; n < 10; ++n) > + for (n = 0; n < num_digits_len; ++n) > { > /* Get the string for the digits with value N. */ > #ifdef COMPILE_WSCANF > @@ -1553,7 +1563,7 @@ __vfscanf_internal (FILE *s, const char *format, va_list argptr, > DIAG_IGNORE_NEEDS_COMMENT (4.7, "-Wmaybe-uninitialized"); > > if (__glibc_unlikely (map != NULL)) > - wcdigits[n] = wcdigits_extended[n]; > + wcdigits[n] = digits_extended[n]; > else > wcdigits[n] = (const wchar_t *) > _NL_CURRENT (LC_CTYPE, _NL_CTYPE_INDIGITS0_WC + n); > @@ -1574,7 +1584,7 @@ __vfscanf_internal (FILE *s, const char *format, va_list argptr, > int avail = width > 0 ? width : INT_MAX; > > if (__glibc_unlikely (map != NULL)) > - mbdigits[n] = mbdigits_extended[n]; > + mbdigits[n] = digits_extended[n]; > else > mbdigits[n] > = curctype->values[_NL_CTYPE_INDIGITS0_MB + n].string; > @@ -1617,13 +1627,13 @@ __vfscanf_internal (FILE *s, const char *format, va_list argptr, > #endif > } > > - if (n == 10) > + if (n == num_digits_len) > { > /* Have not yet found the digit. */ > for (level = from_level + 1; level <= to_level; ++level) > { > /* Search all ten digits of this level. */ > - for (n = 0; n < 10; ++n) > + for (n = 0; n < num_digits_len; ++n) > { > #ifdef COMPILE_WSCANF > if (c == (wint_t) *wcdigits[n]) > @@ -1679,7 +1689,7 @@ __vfscanf_internal (FILE *s, const char *format, va_list argptr, > } > } > > - if (n < 10) > + if (n < num_digits_len) > c = L_('0') + n; > else if (flags & GROUP) > { > @@ -1708,7 +1718,7 @@ __vfscanf_internal (FILE *s, const char *format, va_list argptr, > { > __set_errno (ENOMEM); > done = EOF; > - goto errout; > + break; > } > > if (*cmpp != '\0') > @@ -1742,6 +1752,13 @@ __vfscanf_internal (FILE *s, const char *format, va_list argptr, > > c = inchar (); > } > + > +digits_extended_fail: > + for (n = 0; n < num_digits_len; n++) > + free (digits_extended[n]); > + > + if (done == EOF) > + goto errout; > } > else > /* Read the number into workspace. */ > diff --git a/wcsmbs/Makefile b/wcsmbs/Makefile > index 22192985e1..ed2660a524 100644 > --- a/wcsmbs/Makefile > +++ b/wcsmbs/Makefile > @@ -175,6 +175,7 @@ tests := \ > tst-wscanf-binary-c2x \ > tst-wscanf-binary-gnu11 \ > tst-wscanf-binary-gnu89 \ > + tst-wscanf-to_inpunct \ > wcsatcliff \ > wcsmbs-tst1 \ > # tests > @@ -186,6 +187,7 @@ LOCALES := \ > de_DE.ISO-8859-1 \ > de_DE.UTF-8 \ > en_US.ANSI_X3.4-1968 \ > + fa_IR.UTF-8 \ > hr_HR.ISO-8859-2 \ > ja_JP.EUC-JP \ > tr_TR.ISO-8859-9 \ > @@ -207,6 +209,7 @@ $(objpfx)tst-c16-surrogate.out: $(gen-locales) > $(objpfx)tst-c32-state.out: $(gen-locales) > $(objpfx)test-c8rtomb.out: $(gen-locales) > $(objpfx)test-mbrtoc8.out: $(gen-locales) > +$(objpfx)tst-wscanf-to_inpunct.out: $(gen-locales) > endif > > $(objpfx)tst-wcstod-round: $(libm) > diff --git a/wcsmbs/tst-wscanf-to_inpunct.c b/wcsmbs/tst-wscanf-to_inpunct.c > new file mode 100644 > index 0000000000..72f2a1a422 > --- /dev/null > +++ b/wcsmbs/tst-wscanf-to_inpunct.c > @@ -0,0 +1,79 @@ > +/* Test scanf for languages with mapping pairs of alternate digits and > + separators. > + Copyright (C) 2023 Free Software Foundation, Inc. > + This file is part of the GNU C Library. > + > + The GNU C Library is free software; you can redistribute it and/or > + modify it under the terms of the GNU Lesser General Public > + License as published by the Free Software Foundation; either > + version 2.1 of the License, or (at your option) any later version. > + > + The GNU C Library is distributed in the hope that it will be useful, > + but WITHOUT ANY WARRANTY; without even the implied warranty of > + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU > + Lesser General Public License for more details. > + > + You should have received a copy of the GNU Lesser General Public > + License along with the GNU C Library; if not, see > + . */ > + > +#include > +#include > +#include > +#include > +#include > + > +/* fa_IR defines to_inpunct for numbers. */ > +static const struct input_t > +{ > + int n; > + const wchar_t str[5]; > +} inputs[] = > +{ > + { 1, { 0x000006f1, L'\0' } }, > + { 2, { 0x000006f2, L'\0' } }, > + { 3, { 0x000006f3, L'\0' } }, > + { 4, { 0x000006f4, L'\0' } }, > + { 5, { 0x000006f5, L'\0' } }, > + { 6, { 0x000006f6, L'\0' } }, > + { 7, { 0x000006f7, L'\0' } }, > + { 8, { 0x000006f8, L'\0' } }, > + { 9, { 0x000006f9, L'\0' } }, > + { 10, { 0x000006f1, 0x000006f0, L'\0' } }, > + { 11, { 0x000006f1, 0x000006f1, L'\0' } }, > + { 12, { 0x000006f1, 0x000006f2, L'\0' } }, > + { 13, { 0x000006f1, 0x000006f3, L'\0' } }, > + { 14, { 0x000006f1, 0x000006f4, L'\0' } }, > + { 15, { 0x000006f1, 0x000006f5, L'\0' } }, > + { 16, { 0x000006f1, 0x000006f6, L'\0' } }, > + { 17, { 0x000006f1, 0x000006f7, L'\0' } }, > + { 18, { 0x000006f1, 0x000006f8, L'\0' } }, > + { 19, { 0x000006f1, 0x000006f9, L'\0' } }, > + { 20, { 0x000006f2, 0x000006f0, L'\0' } }, > + { 30, { 0x000006f3, 0x000006f0, L'\0' } }, > + { 40, { 0x000006f4, 0x000006f0, L'\0' } }, > + { 50, { 0x000006f5, 0x000006f0, L'\0' } }, > + { 60, { 0x000006f6, 0x000006f0, L'\0' } }, > + { 70, { 0x000006f7, 0x000006f0, L'\0' } }, > + { 80, { 0x000006f8, 0x000006f0, L'\0' } }, > + { 90, { 0x000006f9, 0x000006f0, L'\0' } }, > + { 100, { 0x000006f1, 0x000006f0, 0x000006f0, L'\0' } }, > + { 1000, { 0x000006f1, 0x000006f0, 0x000006f0, 0x000006f0, L'\0' } }, > +}; > + > +static int > +do_test (void) > +{ > + xsetlocale (LC_ALL, "fa_IR.UTF-8"); > + > + for (int i = 0; i < array_length (inputs); i++) > + { > + int n; > + swscanf (inputs[i].str, L"%Id", &n); > + TEST_COMPARE (n, inputs[i].n); > + } > + > + return 0; > +} > + > +#include > -- > 2.34.1 >