From mboxrd@z Thu Jan 1 00:00:00 1970 Delivery-date: Thu, 19 Feb 2026 12:44:11 +0100 Received: from metis.whiteo.stw.pengutronix.de ([2a0a:edc0:2:b01:1d::104]) by lore.white.stw.pengutronix.de with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.96) (envelope-from ) id 1vt2S2-003OfA-0V for lore@lore.pengutronix.de; Thu, 19 Feb 2026 12:44:11 +0100 Received: from bombadil.infradead.org ([2607:7c80:54:3::133]) by metis.whiteo.stw.pengutronix.de with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1vt2S1-0004gY-Vu for lore@pengutronix.de; Thu, 19 Feb 2026 12:44:10 +0100 DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20210309; h=Sender:Cc:List-Subscribe: List-Help:List-Post:List-Archive:List-Unsubscribe:List-Id: Content-Transfer-Encoding:Content-Type:In-Reply-To:From:References:To:Subject :MIME-Version:Date:Message-ID:Reply-To:Content-ID:Content-Description: Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID: List-Owner; bh=J/YU+TUTlpSye7uH/ZGyT0kKvC37dDK6/dkMZ2rUfMw=; b=i0YvnucnVQ4P43 rRqzPeW+JtKJ5qaTxlPGA4UiUqBIe3a8yDt2CAH1R9Y7o5MTyYeDfW4t0aezTnyPYuHySbOVc92Dn UuP2YkFSUj/795wfNgzzb2gM4N1BRT5KaPpnbkYdTohAkRT3JF7x1TCU7+MuVrSubiQvM/IHY+mKW LHh9X5pUp0Wr7nGq6TGP0ApqxB1pRYXPF8K65M/d2ISZwu1u+hAafWiZGcQL6hhL9xxPjyuhzTlJJ h71+IQnTsfvyBjq5HtKR1XzKZmClwPDs4QYm8AZriLuJqqsH+nCnScck0zTXbMctvKun0esCf3kn4 b/nYb27CUoFZxPmdYvAg==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.98.2 #2 (Red Hat Linux)) id 1vt2RL-0000000BIZT-43Jc; Thu, 19 Feb 2026 11:43:27 +0000 Received: from metis.whiteo.stw.pengutronix.de ([2a0a:edc0:2:b01:1d::104]) by bombadil.infradead.org with esmtps (Exim 4.98.2 #2 (Red Hat Linux)) id 1vt2RJ-0000000BIZ6-2SRA for barebox@lists.infradead.org; Thu, 19 Feb 2026 11:43:27 +0000 Received: from ptz.office.stw.pengutronix.de ([2a0a:edc0:0:900:1d::77] helo=[127.0.0.1]) by metis.whiteo.stw.pengutronix.de with esmtp (Exim 4.92) (envelope-from ) id 1vt2RH-0004cO-RV; Thu, 19 Feb 2026 12:43:23 +0100 Message-ID: <10ed09f9-6a16-4710-9024-0cf1a82eadf5@pengutronix.de> Date: Thu, 19 Feb 2026 12:43:23 +0100 MIME-Version: 1.0 User-Agent: Mozilla Thunderbird To: Marco Felsch References: <20260218-keytoc-multi-env-v2-0-3ea146c95d18@pengutronix.de> <20260218-keytoc-multi-env-v2-1-3ea146c95d18@pengutronix.de> From: Jonas Rebmann Content-Language: en-US In-Reply-To: Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 7bit X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20260219_034325_789763_1C627089 X-CRM114-Status: GOOD ( 39.35 ) X-BeenThere: barebox@lists.infradead.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: BAREBOX , Ahmad Fatoum Sender: "barebox" X-SA-Exim-Connect-IP: 2607:7c80:54:3::133 X-SA-Exim-Mail-From: barebox-bounces+lore=pengutronix.de@lists.infradead.org X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on metis.whiteo.stw.pengutronix.de X-Spam-Level: X-Spam-Status: No, score=-3.3 required=4.0 tests=AWL,BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_NONE autolearn=unavailable autolearn_force=no version=3.4.2 Subject: Re: [PATCH v2 1/2] crypto: keytoc: Split env-provided full keyspec on spaces X-SA-Exim-Version: 4.2.1 (built Wed, 08 May 2019 21:11:16 +0000) X-SA-Exim-Scanned: Yes (on metis.whiteo.stw.pengutronix.de) Hi Marco On 2026-02-18 13:55, Marco Felsch wrote: > On 26-02-18, Jonas Rebmann wrote: >> keytoc/CONFIG_CRYPTO_PUBLIC_KEYS can work with a complete keyspec >> provided by an environment variable as opposed to providing single URIs. >> This would be a very useful feature if it could also provide any number >> of keys. Kconfig however provides keytoc with regular keyspecs already >> split at spaces so without furhter measures, the env variable can only >> be expanded into a single key. >> >> If a complete argument is provided via __ENV, split it at any space >> character that is not escaped with a backslash in front of it. An >> actual backslash in a path needs to be escape with another backslash. >> >> Signed-off-by: Jonas Rebmann >> --- >> scripts/include/linux/string.h | 2 +- >> scripts/include/string_util.h | 65 ++++++++++++++++++++++++++++++++++++++ >> scripts/keytoc.c | 71 +++++++++++++++++++++++++++++------------- > > Porting strsep_unescaped should be one patch and the keytoc changes > should be one patch. I will send a v3 with that change. >> 3 files changed, 115 insertions(+), 23 deletions(-) >> >> diff --git a/scripts/include/linux/string.h b/scripts/include/linux/string.h >> index 649287b80a..ac06f96d8d 100644 >> --- a/scripts/include/linux/string.h >> +++ b/scripts/include/linux/string.h >> @@ -16,4 +16,4 @@ int strtobool(const char *s, bool *res); >> extern size_t strlcpy(char *dest, const char *src, size_t size); >> #endif >> >> -#endif /* _LINUX_STRING_H_ */ >> +#endif /* _TOOLS_LINUX_STRING_H_ */ >> diff --git a/scripts/include/string_util.h b/scripts/include/string_util.h >> new file mode 100644 >> index 0000000000..e71aa60d26 >> --- /dev/null >> +++ b/scripts/include/string_util.h >> @@ -0,0 +1,65 @@ >> +#ifndef _TOOLS_STRING_UTIL_H_ >> +#define _TOOLS_STRING_UTIL_H_ >> + >> +#include >> +#include >> + >> +// SPDX-SnippetBegin >> +// SPDX-Snippet-Comment: Origin-URL: https://git.pengutronix.de/cgit/barebox/tree/lib/string.c?id=dfcf686f94a5a5387660f2afab79a714baab828a >> + >> +/** >> + * strsep_unescaped - Split a string into tokens, while ignoring escaped delimiters >> + * @s: The string to be searched >> + * @ct: The delimiter characters to search for >> + * @delim: optional pointer to store found delimiter into >> + * >> + * strsep_unescaped() behaves like strsep unless it meets an escaped delimiter. >> + * In that case, it shifts the string back in memory to overwrite the escape's >> + * backslash then continues the search until an unescaped delimiter is found. >> + * >> + * On end of string, this function returns NULL. As long as a non-NULL >> + * value is returned and @delim is not NULL, the found delimiter will >> + * be stored into *@delim. >> + */ >> +static char *strsep_unescaped(char **s, const char *ct, char *delim) >> +{ >> + char *sbegin = *s, *hay; >> + const char *needle; >> + size_t shift = 0; >> + >> + if (sbegin == NULL) >> + return NULL; >> + >> + for (hay = sbegin; *hay != '\0'; ++hay) { >> + *hay = hay[shift]; >> + >> + if (*hay == '\\') { >> + *hay = hay[++shift]; >> + if (*hay != '\\') >> + continue; >> + } >> + >> + for (needle = ct; *needle != '\0'; ++needle) { >> + if (*hay == *needle) >> + goto match; >> + } >> + } >> + >> + *s = NULL; >> + if (delim) >> + *delim = '\0'; >> + return sbegin; >> + >> +match: >> + if (delim) >> + *delim = *hay; >> + *hay = '\0'; >> + *s = &hay[shift + 1]; >> + >> + return sbegin; >> +} >> + >> +// SPDX-SnippetEnd >> + >> + >> +#endif /* _TOOLS_STRING_UTIL_H_ */ >> diff --git a/scripts/keytoc.c b/scripts/keytoc.c >> index 77ada3af45..ed091285f4 100644 >> --- a/scripts/keytoc.c >> +++ b/scripts/keytoc.c >> @@ -6,9 +6,12 @@ >> * URI to a C struct suitable to compile with barebox. >> * >> * TODO: Find a better way for reimport_key() >> - * >> */ >> -#pragma GCC diagnostic ignored "-Wdeprecated-declarations" >> + >> +#pragma GCC diagnostic ignored "-Wdeprecated-declarations" /* ENGINE deprecated in OpenSSL 3.0 */ This line is admittedly unrelated to what I did but I think our convention here is to rather slip such tiny things into whereever they come up than to create an extra patch for it. Correct me if I'm wrong. >> + >> +#include "include/string_util.h" >> + >> #include >> #include >> #include >> @@ -784,8 +787,13 @@ static bool parse_info(char *p, struct keyinfo *out) >> } >> } >> >> -static bool get_name_path(const char *keyspec, struct keyinfo *out) >> +static bool parse_keyspec(const char *keyspec, struct keyinfo *out) >> { >> + if (!strncmp(keyspec, "pkcs11:", 7)) { /* legacy format of pkcs11 URI */ >> + out->path = strdup(keyspec); >> + return true; >> + } >> + >> char *sep, *spec; >> >> spec = strdup(keyspec); >> @@ -814,10 +822,10 @@ static bool get_name_path(const char *keyspec, struct keyinfo *out) >> >> int main(int argc, char *argv[]) >> { >> - int i, opt, ret; >> + int argi, opt, ret; > > The diff gets quite hard to read, if you do such changes in between. > Please reduce the changes per patch to the bare minimum. Renaming could > be done later on in a separate patch. The rename is part of the same change. Since environment variables can now be expanded into any number of new keyspecs we're moving from looping arguments as keys to a nested loop over each arguments expansions set of multiple keys. i is a fine name for an index in a loop as long as there's only one index and one loop. Replacing that simple loop with a nested loop is the core of the change I'm making and the rename from i to argi is part of that. >> char *outfile = NULL; >> - int keycount; >> - struct keyinfo *keylist; >> + size_t keycount, num_positionals; >> + struct keyinfo *keylist = NULL; >> >> outfilep = stdout; >> >> @@ -852,22 +860,41 @@ int main(int argc, char *argv[]) >> exit(1); >> } >> >> - keycount = argc - optind; >> - keylist = calloc(sizeof(struct keyinfo), keycount); >> >> - for (i = 0; i < keycount; i++) { >> - const char *keyspec = try_resolve_env(argv[optind + i]); >> - struct keyinfo *info = &keylist[i]; >> + num_positionals = argc - optind; >> + keycount = num_positionals; >> >> - if (!keyspec) >> - exit(1); >> + keylist = calloc(keycount, sizeof(*keylist)); > > Same here, why can't you keep the keycount and keylist handling > unchanged? The same change: As a direct consequence of expanding multiple keys from one environment variable, we cannot just assume that the keylist length is the number of positional arguments, we cannot know the number of keys we end up with in advance and hence we need to dynamically resize the keylist. >> + >> + if (!keylist) >> + enomem_exit("push"); > > This would go into separate patch as well. > > As said above, please split your patchset into multiple smaller patches > to make it easier to read. Those changes to keytoc are mostly one single atomic change that cannot be split into more patches. This includes new memory allocations to handle growth of the keylist. > Regards, > Marco > > >> + >> + int listi = 0; >> + >> + for (argi = 0; argi < num_positionals; argi++) { >> + char *arg = strdup(argv[optind + argi]); >> + char *resolved = try_resolve_env(arg); >> >> - if (!strncmp(keyspec, "pkcs11:", 7)) { // legacy format of pkcs11 URI >> - info->path = strdup(keyspec); >> + if (arg == resolved) { >> + keylist[listi].path = arg; >> + listi++; >> } else { >> - if (!get_name_path(keyspec, info)) { >> - fprintf(stderr, "invalid keyspec %i: %s\n", optind, keyspec); >> - exit(1); >> + char *keyspecs = strdup(resolved); >> + char *keyspec; >> + >> + keycount--; >> + while ((keyspec = strsep_unescaped(&keyspecs, " ", NULL))) { >> + keycount++; >> + keylist = reallocarray(keylist, keycount, sizeof(*keylist)); >> + if (!keylist) >> + enomem_exit("realloc keylist"); >> + bzero(keylist + (keycount - 1), sizeof(*keylist)); >> + if (!parse_keyspec(keyspec, &keylist[listi])) { >> + fprintf(stderr, "invalid keyspec %i: %s\n", optind, >> + keyspec); >> + exit(1); >> + } >> + listi++; >> } >> } >> } >> @@ -885,14 +912,14 @@ int main(int argc, char *argv[]) >> } >> >> >> - for (i = 0; i < keycount; i++) { >> - struct keyinfo *info = &keylist[i]; >> + for (argi = 0; argi < keycount; argi++) { >> + struct keyinfo *info = &keylist[argi]; >> >> /* resolve __ENV__ for name_hint and path */ >> info->name_hint = try_resolve_env(info->name_hint); >> info->path = try_resolve_env(info->path); >> >> - if (asprintf(&info->name_c, "key_%i", i + 1) < 0) >> + if (asprintf(&info->name_c, "key_%i", argi + 1) < 0) >> enomem_exit("asprintf"); >> >> /* unfortunately, the fit name hint is mandatory in the barebox codebase */ >> @@ -901,7 +928,7 @@ int main(int argc, char *argv[]) >> >> if (!info->keyring) { >> info->keyring = strdup("fit"); >> - fprintf(stderr, "Warning: No keyring provided in keyspec, defaulting to keyring=fit for %s\n", argv[optind + i]); >> + fprintf(stderr, "Warning: No keyring provided in keyspec, defaulting to keyring=fit for %s\n", argv[optind + argi]); >> } >> >> ret = gen_key(info); >> >> -- >> 2.51.2.535.g419c72cb8a >> >> >> > -- Pengutronix e.K. | Jonas Rebmann | Steuerwalder Str. 21 | http://www.pengutronix.de/ | 31137 Hildesheim, Germany | Phone: +49-5121-206917-0 | Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-9 |