- gerrit-log - lists.osmocom.org

[M] Change in pysim[master]: saip.PES.rebuild_mandatory_services(): set 5G get-identity, profile-a...
by neels 12 Jun '26

12 Jun '26

neels has uploaded this change for review. ( https://gerrit.osmocom.org/c/pysim/+/42829?usp=email ) Change subject: saip.PES.rebuild_mandatory_services(): set 5G get-identity, profile-a-x25519, profile-b-p256 ...................................................................... saip.PES.rebuild_mandatory_services(): set 5G get-identity, profile-a-x25519, profile-b-p256 Change-Id: Ibc29c6437c5c92e2b14938b733156536863465c1 --- M pySim/esim/saip/__init__.py 1 file changed, 51 insertions(+), 2 deletions(-) git pull ssh://gerrit.osmocom.org:29418/pysim refs/changes/29/42829/1 diff --git a/pySim/esim/saip/__init__.py b/pySim/esim/saip/__init__.py index ec59c50..5fe8136 100644 --- a/pySim/esim/saip/__init__.py +++ b/pySim/esim/saip/__init__.py @@ -34,7 +34,7 @@ from pySim.utils import dec_imsi from pySim.ts_102_221 import FileDescriptor from pySim.filesystem import CardADF, Path -from pySim.ts_31_102 import ADF_USIM +from pySim.ts_31_102 import ADF_USIM, EF_UST, EF_SUCI_Calc_Info from pySim.ts_31_103 import ADF_ISIM from pySim.esim import compile_asn1_subdir from pySim.esim.saip import templates @@ -1726,7 +1726,56 @@ if 'BT' in ftype_list: svc_set.add('ber-tlv') # FIXME:dfLinked files (scan all files, check for non-empty Fcp.linkPath presence of DFs) - # TODO: 5G related bits (derive from EF.UST or file presence?) + + # 5G: + # - When SUCI is: + # - enabled (EF.UST 124 = true) + # AND + # - calculated in the USIM (EF.UST 125 = true), + # then eUICC-Mandatory-services needs 'get-identity'. + # - 'get-identity' implies that the eUICC must support ONE OF profile-A OR profile-B. + # So, when SUCI-CalcInfo for USIM in DF.SAIP contains both key types, + # then no profile-A or B services need to be requested explicitly. + # - When the SUCI-CalcInfo for USIM (DF.SAIP) contains ONLY a key of profile-A ("identifier": 1), + # then eUICC-Mandatory-services needs 'profile-a-x25519'. + # - Same: ONLY profile-B ("identifier": 2) needs 'profile-b-p256'. + # - (When SUCI is calculated in the UE, then the eUICC does not need to provide any of these services.) + suci_in_usim_enabled = False + try: + f_ust = self.get_pe_for_type("usim").files["ef-ust"] + ust = EF_UST().decode_bin(f_ust.body) + if ust[124]['activated'] and ust[125]['activated']: + suci_in_usim_enabled = True + except (KeyError, AttributeError): + pass + if suci_in_usim_enabled: + svc_set.add('get-identity') + # now check for profile-a and profile-b + suci_calcinfo_has_profile_a = False + suci_calcinfo_has_profile_b = False + try: + f_sucici = self.get_pe_for_type("df-saip").files["ef-suci-calc-info-usim"] + sucici = EF_SUCI_Calc_Info().decode_bin(f_sucici.body) or {} + for prot_scheme in sucici['prot_scheme_id_list']: + if not isinstance(prot_scheme, dict): + continue + ps_id = prot_scheme["identifier"] + if ps_id == 1: + suci_calcinfo_has_profile_a = True + elif ps_id == 2: + suci_calcinfo_has_profile_b = True + except (KeyError, AttributeError): + pass + if suci_calcinfo_has_profile_a and suci_calcinfo_has_profile_b: + # 'get-identity' implies that the eUICC supports one of the above. Do not require a specific one. + pass + elif suci_calcinfo_has_profile_a: + # The profile has only a profile-A key, so require that + svc_set.add('profile-a-x25519') + elif suci_calcinfo_has_profile_b: + # The profile has only a profile-B key, so require that + svc_set.add('profile-b-p256') + hdr_pe = self.get_pe_for_type('header') # patch in the 'manual' services from the existing list: for old_svc in hdr_pe.decoded['eUICC-Mandatory-services'].keys(): -- To view, visit https://gerrit.osmocom.org/c/pysim/+/42829?usp=email To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings?usp=email Gerrit-MessageType: newchange Gerrit-Project: pysim Gerrit-Branch: master Gerrit-Change-Id: Ibc29c6437c5c92e2b14938b733156536863465c1 Gerrit-Change-Number: 42829 Gerrit-PatchSet: 1 Gerrit-Owner: neels <nhofmeyr(a)sysmocom.de>

1 0

[M] Change in osmo-ccid-firmware[master]: libosmo_emb: type-safe tearfree_u64_t wrapper for LDRD/STRD access
by laforge 12 Jun '26

12 Jun '26

laforge has submitted this change. ( https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42784?usp=email ) Change subject: libosmo_emb: type-safe tearfree_u64_t wrapper for LDRD/STRD access ...................................................................... libosmo_emb: type-safe tearfree_u64_t wrapper for LDRD/STRD access Although types are frowned upon because memorizing all differerent flavors of void* is the usual way to get acquainted with any mature C code base some heretics decided to introduce generics in C11, which can be used to make aligned access (which guarantees tear-free/restartable 64 bit access on cortex m4) less exciting. Change-Id: I458bfaf53e27c51882c7a8c1326e51d12943b0bb --- M ccid_common/cuart.h M ccid_host/libosmo_emb.h M sysmoOCTSIM/libosmo_emb.c M sysmoOCTSIM/libosmo_emb.h 4 files changed, 45 insertions(+), 20 deletions(-) Approvals: jolly: Looks good to me, but someone else must approve laforge: Looks good to me, approved; Verified diff --git a/ccid_common/cuart.h b/ccid_common/cuart.h index a53dc18..d31d2d9 100644 --- a/ccid_common/cuart.h +++ b/ccid_common/cuart.h @@ -102,7 +102,7 @@ uint32_t wtime_etu; /* deadline in jiffies (ms) for card response timeout, 0 = inactive. * Set from IRQ context, checked from main loop */ - volatile uint64_t wtime_deadline __attribute__((aligned(8))); + tearfree_u64_t wtime_deadline; /* expected number of bytes, for timeout */ uint32_t current_wtime_byte; diff --git a/ccid_host/libosmo_emb.h b/ccid_host/libosmo_emb.h index ffc0324..e811dd3 100644 --- a/ccid_host/libosmo_emb.h +++ b/ccid_host/libosmo_emb.h @@ -1,15 +1,23 @@ #pragma once -/* Host-side jiffies emulation */ +/* Host-side jiffies emulation + * On the host every access is a plain memory access; the typed wrapper exists + * only purely for type compatibility with code shared with the embedded target. */ #include <stdint.h> +#include <stdalign.h> -static inline uint64_t ldrd_u64(const volatile uint64_t *p) -{ - return *p; -} -static inline void strd_u64(volatile uint64_t *p, uint64_t v) -{ - *p = v; -} +typedef struct { + alignas(8) volatile uint64_t _v; +} tearfree_u64_t; + +static inline uint64_t _ldrd_u64(const tearfree_u64_t *p) { return p->_v; } +static inline void _strd_u64(tearfree_u64_t *p, uint64_t v) { p->_v = v; } + +#define ldrd_u64(p) _Generic((p), \ + tearfree_u64_t *: _ldrd_u64, \ + const tearfree_u64_t *: _ldrd_u64)(p) + +#define strd_u64(p, v) _Generic((p), \ + tearfree_u64_t *: _strd_u64)((p), (v)) uint64_t get_jiffies(void); diff --git a/sysmoOCTSIM/libosmo_emb.c b/sysmoOCTSIM/libosmo_emb.c index e96f88e..731268b 100644 --- a/sysmoOCTSIM/libosmo_emb.c +++ b/sysmoOCTSIM/libosmo_emb.c @@ -17,7 +17,7 @@ #include "driver_init.h" #include "libosmo_emb.h" -volatile uint64_t jiffies __attribute__((aligned(8))); +tearfree_u64_t jiffies; uint64_t get_jiffies(void) { @@ -31,7 +31,7 @@ void SysTick_Handler(void) { - jiffies++; + jiffies._v++; } int _gettimeofday(struct timeval *tv, void *tz) diff --git a/sysmoOCTSIM/libosmo_emb.h b/sysmoOCTSIM/libosmo_emb.h index 4b576c6..705982c 100644 --- a/sysmoOCTSIM/libosmo_emb.h +++ b/sysmoOCTSIM/libosmo_emb.h @@ -7,42 +7,59 @@ * * Worst-case outcome is being off by one tick (1 ms) which is irrelevant * for the timeout use-case. + * + * To enforce the alignment requirement the underlying + * uint64_t is wrapped in a struct with 8-byte + * alignment via alignas(8) + _Generic to ensure types. */ #include <stdint.h> +#include <stdalign.h> + +typedef struct { + alignas(8) volatile uint64_t _v; +} tearfree_u64_t; #if defined(__arm__) -static inline uint64_t ldrd_u64(const volatile uint64_t *p) +static inline uint64_t _ldrd_u64(const tearfree_u64_t *p) { uint32_t lo, hi; __asm__ volatile( "ldrd %0, %1, [%2]\n" : "=&r"(lo), "=&r"(hi) - : "r"(p) + : "r"(&p->_v) : "memory"); return ((uint64_t)hi << 32) | lo; } -static inline void strd_u64(volatile uint64_t *p, uint64_t v) +static inline void _strd_u64(tearfree_u64_t *p, uint64_t v) { uint32_t lo = (uint32_t)v; uint32_t hi = (uint32_t)(v >> 32); __asm__ volatile( "strd %1, %2, [%0]\n" - : : "r"(p), "r"(lo), "r"(hi) + : : "r"(&p->_v), "r"(lo), "r"(hi) : "memory"); } #else /* host */ -static inline uint64_t ldrd_u64(const volatile uint64_t *p) +static inline uint64_t _ldrd_u64(const tearfree_u64_t *p) { - return *p; + return p->_v; } -static inline void strd_u64(volatile uint64_t *p, uint64_t v) +static inline void _strd_u64(tearfree_u64_t *p, uint64_t v) { - *p = v; + p->_v = v; } #endif +/* _Generic ensures matched types or bust */ +#define ldrd_u64(p) _Generic((p), \ + tearfree_u64_t *: _ldrd_u64, \ + const tearfree_u64_t *: _ldrd_u64)(p) + +#define strd_u64(p, v) _Generic((p), \ + tearfree_u64_t *: _strd_u64)((p), (v)) + uint64_t get_jiffies(void); void store_jiffies(uint64_t j); -- To view, visit https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42784?usp=email To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings?usp=email Gerrit-MessageType: merged Gerrit-Project: osmo-ccid-firmware Gerrit-Branch: master Gerrit-Change-Id: I458bfaf53e27c51882c7a8c1326e51d12943b0bb Gerrit-Change-Number: 42784 Gerrit-PatchSet: 3 Gerrit-Owner: Hoernchen <ewild(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: jolly <andreas(a)eversberg.eu> Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: lynxis lazus <lynxis(a)fe80.eu>

1 0

[M] Change in osmo-ccid-firmware[master]: libosmo_emb: type-safe tearfree_u64_t wrapper for LDRD/STRD access
by laforge 12 Jun '26

12 Jun '26

Attention is currently required from: Hoernchen, jolly, lynxis lazus. laforge has removed a vote from this change. ( https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42784?usp=email ) Change subject: libosmo_emb: type-safe tearfree_u64_t wrapper for LDRD/STRD access ...................................................................... Removed Verified-1 by Jenkins Builder (1000002) -- To view, visit https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42784?usp=email To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings?usp=email Gerrit-MessageType: deleteVote Gerrit-Project: osmo-ccid-firmware Gerrit-Branch: master Gerrit-Change-Id: I458bfaf53e27c51882c7a8c1326e51d12943b0bb Gerrit-Change-Number: 42784 Gerrit-PatchSet: 3 Gerrit-Owner: Hoernchen <ewild(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: jolly <andreas(a)eversberg.eu> Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: lynxis lazus <lynxis(a)fe80.eu> Gerrit-Attention: Hoernchen <ewild(a)sysmocom.de> Gerrit-Attention: jolly <andreas(a)eversberg.eu> Gerrit-Attention: lynxis lazus <lynxis(a)fe80.eu>

1 0

[M] Change in osmo-ccid-firmware[master]: libosmo_emb: type-safe tearfree_u64_t wrapper for LDRD/STRD access
by laforge 12 Jun '26

12 Jun '26

Attention is currently required from: Hoernchen, jolly, lynxis lazus. laforge has posted comments on this change by Hoernchen. ( https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42784?usp=email ) Change subject: libosmo_emb: type-safe tearfree_u64_t wrapper for LDRD/STRD access ...................................................................... Patch Set 3: Code-Review+2 Verified+1 -- To view, visit https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42784?usp=email To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings?usp=email Gerrit-MessageType: comment Gerrit-Project: osmo-ccid-firmware Gerrit-Branch: master Gerrit-Change-Id: I458bfaf53e27c51882c7a8c1326e51d12943b0bb Gerrit-Change-Number: 42784 Gerrit-PatchSet: 3 Gerrit-Owner: Hoernchen <ewild(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: jolly <andreas(a)eversberg.eu> Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: lynxis lazus <lynxis(a)fe80.eu> Gerrit-Attention: Hoernchen <ewild(a)sysmocom.de> Gerrit-Attention: jolly <andreas(a)eversberg.eu> Gerrit-Attention: lynxis lazus <lynxis(a)fe80.eu> Gerrit-Comment-Date: Fri, 12 Jun 2026 16:58:23 +0000 Gerrit-HasComments: No Gerrit-Has-Labels: Yes

1 0

[S] Change in osmo-ccid-firmware[master]: 7816fsm: fail PPS on invalid first byte in PPS_S_WAIT_PPSX
by laforge 12 Jun '26

12 Jun '26

laforge has submitted this change. ( https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42780?usp=email ) ( 1 is the latest approved patch-set. No files were changed between the latest approved patch-set and the submitted one. )Change subject: 7816fsm: fail PPS on invalid first byte in PPS_S_WAIT_PPSX ...................................................................... 7816fsm: fail PPS on invalid first byte in PPS_S_WAIT_PPSX Change-Id: I5b74b8443a98224c0c95a664a886066495d8b64a --- M ccid_common/iso7816_fsm.c 1 file changed, 15 insertions(+), 1 deletion(-) Approvals: jolly: Looks good to me, but someone else must approve laforge: Looks good to me, approved Jenkins Builder: Verified diff --git a/ccid_common/iso7816_fsm.c b/ccid_common/iso7816_fsm.c index 2b3b41e..c0ff1f0 100644 --- a/ccid_common/iso7816_fsm.c +++ b/ccid_common/iso7816_fsm.c @@ -1076,8 +1076,22 @@ msgb_put_u8(atp->rx_cmd, byte); switch (fi->state) { case PPS_S_WAIT_PPSX: - if (byte == 0xff) + /* ISO 7816-3 §9.2: PPSS is fixed at 0xff. Any other + * first byte is an erroneous PPS response and §9.1 + * requires deactivation. We must transition out of + * WAIT_PPSX on every byte like every other PPS substate + * so the unconditional msgb_put_u8 above stays + * bounded by the spec's 6-byte maximum. */ + if (byte == 0xff) { osmo_fsm_inst_state_chg(fi, PPS_S_WAIT_PPS0, 0, 0); + } else { + LOGPFSML(fi, LOGL_ERROR, + "Invalid PPSS=0x%02x (expected 0xff); failing PPS\n", + byte); + osmo_fsm_inst_state_chg(fi, PPS_S_DONE, 0, 0); + osmo_fsm_inst_dispatch(fi->proc.parent, + ISO7816_E_PPS_FAILED_IND, atp->tx_cmd); + } break; case PPS_S_WAIT_PPS0: atp->pps0_recv = byte; -- To view, visit https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42780?usp=email To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings?usp=email Gerrit-MessageType: merged Gerrit-Project: osmo-ccid-firmware Gerrit-Branch: master Gerrit-Change-Id: I5b74b8443a98224c0c95a664a886066495d8b64a Gerrit-Change-Number: 42780 Gerrit-PatchSet: 2 Gerrit-Owner: Hoernchen <ewild(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: jolly <andreas(a)eversberg.eu> Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: lynxis lazus <lynxis(a)fe80.eu>

1 0

[XS] Change in osmo-ccid-firmware[master]: ccid: ICC_MUTE instead of CMD_NOT_SUPPORTED on XfrBlock to unpowered ...
by laforge 12 Jun '26

12 Jun '26

laforge has submitted this change. ( https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42783?usp=email ) ( 1 is the latest approved patch-set. No files were changed between the latest approved patch-set and the submitted one. )Change subject: ccid: ICC_MUTE instead of CMD_NOT_SUPPORTED on XfrBlock to unpowered slot ...................................................................... ccid: ICC_MUTE instead of CMD_NOT_SUPPORTED on XfrBlock to unpowered slot The command is supported, just currently impossible. Closes:OS#7015 Change-Id: I7f64475b023bd2b6fd1c4263850e56dd84d20b3e --- M ccid_common/ccid_slot_fsm.c 1 file changed, 1 insertion(+), 1 deletion(-) Approvals: jolly: Looks good to me, but someone else must approve Jenkins Builder: Verified laforge: Looks good to me, approved diff --git a/ccid_common/ccid_slot_fsm.c b/ccid_common/ccid_slot_fsm.c index fae2c3b..d680994 100644 --- a/ccid_common/ccid_slot_fsm.c +++ b/ccid_common/ccid_slot_fsm.c @@ -370,7 +370,7 @@ /* might be unpowered after failed ppss that led to reset */ if (cs->icc_powered != true) - return -0; + return -CCID_ERR_ICC_MUTE; msgb_pull(msg, 10); -- To view, visit https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42783?usp=email To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings?usp=email Gerrit-MessageType: merged Gerrit-Project: osmo-ccid-firmware Gerrit-Branch: master Gerrit-Change-Id: I7f64475b023bd2b6fd1c4263850e56dd84d20b3e Gerrit-Change-Number: 42783 Gerrit-PatchSet: 2 Gerrit-Owner: Hoernchen <ewild(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: jolly <andreas(a)eversberg.eu> Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: lynxis lazus <lynxis(a)fe80.eu>

1 0

[XS] Change in osmo-ccid-firmware[master]: 7816fsm: state_chg before card_uart_tx in tpdu_s_procedure_action
by laforge 12 Jun '26

12 Jun '26

laforge has submitted this change. ( https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42781?usp=email ) Change subject: 7816fsm: state_chg before card_uart_tx in tpdu_s_procedure_action ...................................................................... 7816fsm: state_chg before card_uart_tx in tpdu_s_procedure_action State change must precede the TX to ensure we don't lose TX_COMPLETE notifications that arrive at a bad time. Change-Id: Id66f836452674209e25dc3ee4c37d616de30001b --- M ccid_common/iso7816_fsm.c 1 file changed, 2 insertions(+), 2 deletions(-) Approvals: jolly: Looks good to me, but someone else must approve laforge: Looks good to me, approved Jenkins Builder: Verified diff --git a/ccid_common/iso7816_fsm.c b/ccid_common/iso7816_fsm.c index c0ff1f0..4e4c2b1 100644 --- a/ccid_common/iso7816_fsm.c +++ b/ccid_common/iso7816_fsm.c @@ -1371,8 +1371,8 @@ delay_us(1); gpio_set_pin_level(PIN_PB12, false); #endif - card_uart_tx(ip->uart, msgb_l2(tfp->tpdu), msgb_l2len(tfp->tpdu), true); osmo_fsm_inst_state_chg(fi, TPDU_S_TX_REMAINING, 0, 0); + card_uart_tx(ip->uart, msgb_l2(tfp->tpdu), msgb_l2len(tfp->tpdu), true); } else { /* 7816-3 10.3.2 special case outgoing transfer 0 means 256 */ int len_expected = tpduh->p3 == 0 ? 256 : tpduh->p3; @@ -1391,8 +1391,8 @@ if (tfp->is_command) { /* transmit *next*, not first byte */ OSMO_ASSERT(msgb_l3len(tfp->tpdu) >= 0); - card_uart_tx(ip->uart, msgb_l3(tfp->tpdu), 1, false); osmo_fsm_inst_state_chg(fi, TPDU_S_TX_SINGLE, 0, 0); + card_uart_tx(ip->uart, msgb_l3(tfp->tpdu), 1, false); } else { card_uart_set_rx_threshold(ip->uart, 1); card_uart_ctrl(ip->uart, CUART_CTL_RX_TIMER_HINT, 1); -- To view, visit https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42781?usp=email To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings?usp=email Gerrit-MessageType: merged Gerrit-Project: osmo-ccid-firmware Gerrit-Branch: master Gerrit-Change-Id: Id66f836452674209e25dc3ee4c37d616de30001b Gerrit-Change-Number: 42781 Gerrit-PatchSet: 2 Gerrit-Owner: Hoernchen <ewild(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: jolly <andreas(a)eversberg.eu> Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: lynxis lazus <lynxis(a)fe80.eu>

1 0

[XS] Change in osmo-ccid-firmware[master]: cuart: Fix waiting time to be per-byte instead of total timeout
by laforge 12 Jun '26

12 Jun '26

laforge has submitted this change. ( https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42777?usp=email ) ( 1 is the latest approved patch-set. No files were changed between the latest approved patch-set and the submitted one. )Change subject: cuart: Fix waiting time to be per-byte instead of total timeout ...................................................................... cuart: Fix waiting time to be per-byte instead of total timeout The previous code multiplied WT by the number of expected bytes, creating a total timeout proportional to the transfer size. This works fine for (currently unsupported) high baud rates, but it makes it look like the reader "freezes" at default rates due to the very long delay. Just reset it upon rx and do not multiply it so it behaves as expected. Closes:OS#7012 Change-Id: Ic00040b88e1b204db3f4f3edad09878aa28d35a1 --- M ccid_common/cuart.c M sysmoOCTSIM/cuart_driver_asf4_usart_async.c 2 files changed, 5 insertions(+), 2 deletions(-) Approvals: Jenkins Builder: Verified jolly: Looks good to me, but someone else must approve laforge: Looks good to me, approved diff --git a/ccid_common/cuart.c b/ccid_common/cuart.c index bb17c4b..737a392 100644 --- a/ccid_common/cuart.c +++ b/ccid_common/cuart.c @@ -63,8 +63,9 @@ int etu_in_us = get_etu_in_us(cuart) + 1; cuart->wtime_etu = cuart->wtime_etu ? cuart->wtime_etu : 1; - /* timeout is wtime * ETU * expected number of bytes */ - uint32_t usecs = etu_in_us * cuart->wtime_etu * cuart->current_wtime_byte; + /* ISO 7816-3 Section 10.2: WT is the max delay between consecutive + * characters, not a total transfer timeout. Restart on each byte. */ + uint32_t usecs = etu_in_us * cuart->wtime_etu; /* limit lower wait time to reasonable value */ if (usecs < 300000) diff --git a/sysmoOCTSIM/cuart_driver_asf4_usart_async.c b/sysmoOCTSIM/cuart_driver_asf4_usart_async.c index 42c2110..8330523 100644 --- a/sysmoOCTSIM/cuart_driver_asf4_usart_async.c +++ b/sysmoOCTSIM/cuart_driver_asf4_usart_async.c @@ -32,6 +32,8 @@ int rc; OSMO_ASSERT(cuart); + card_uart_wtime_restart(cuart); + if (cuart->rx_threshold == 1) { /* bypass ringbuffer and report byte directly */ uint8_t rx[1]; -- To view, visit https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42777?usp=email To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings?usp=email Gerrit-MessageType: merged Gerrit-Project: osmo-ccid-firmware Gerrit-Branch: master Gerrit-Change-Id: Ic00040b88e1b204db3f4f3edad09878aa28d35a1 Gerrit-Change-Number: 42777 Gerrit-PatchSet: 2 Gerrit-Owner: Hoernchen <ewild(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: jolly <andreas(a)eversberg.eu> Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: lynxis lazus <lynxis(a)fe80.eu>

1 0

[S] Change in osmo-ccid-firmware[master]: 7816fsm: reset stale cuart state on FSM RESET entry
by laforge 12 Jun '26

12 Jun '26

laforge has submitted this change. ( https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42779?usp=email ) ( 1 is the latest approved patch-set. No files were changed between the latest approved patch-set and the submitted one. )Change subject: 7816fsm: reset stale cuart state on FSM RESET entry ...................................................................... 7816fsm: reset stale cuart state on FSM RESET entry Reset paths reached without power-cycling (WTIME_EXP, HW_ERR, CARD_REMOVAL during a warm reset) leave the cuart with stale tx_busy, rx_threshold and wtime_etu from the prior transaction. The next ATR then hits card_uart_tx tx_busy assertion, or the ATR receive stalls because the 33-byte ATR can never reach a multi-byte rx_threshold left from a TPDU. The new card_uart_tx_abort() clears tx_busy + rx_after_tx_compl + WT, without driving a synthetic TX_COMPLETE through the FSM. iso7816_3_reset_onenter is the right place to do this alongside rx_threshold=1 and wtime_etu=default, this mirrors what card_uart_ctrl(POWER_*=0) already does, but for the warm-reset paths that don't touch power. Change-Id: Iac8bd7f4f0eecccc9acce149277a4f5016fec7c1 --- M ccid_common/ccid_slot_fsm.c M ccid_common/cuart.c M ccid_common/cuart.h M ccid_common/iso7816_fsm.c 4 files changed, 40 insertions(+), 0 deletions(-) Approvals: jolly: Looks good to me, but someone else must approve laforge: Looks good to me, approved Jenkins Builder: Verified diff --git a/ccid_common/ccid_slot_fsm.c b/ccid_common/ccid_slot_fsm.c index 22d3f70..fae2c3b 100644 --- a/ccid_common/ccid_slot_fsm.c +++ b/ccid_common/ccid_slot_fsm.c @@ -54,6 +54,8 @@ static struct iso_fsm_slot_instance g_si; +static void iso_fsm_slot_abort(struct ccid_slot *cs); + static struct iso_fsm_slot *ccid_slot2iso_fsm_slot(struct ccid_slot *cs) { OSMO_ASSERT(cs->slot_nr < ARRAY_SIZE(g_si.slot)); diff --git a/ccid_common/cuart.c b/ccid_common/cuart.c index 737a392..bb47da4 100644 --- a/ccid_common/cuart.c +++ b/ccid_common/cuart.c @@ -167,6 +167,24 @@ return rc; } +/*! Abort any in-flight TX on the cuart side. + * + * Clears tx_busy and rx_after_tx_compl without driving CUART_E_TX_COMPLETE + * through the FSM. Use this when the layer above (the ISO 7816-3 FSM) has + * given up on the current transaction (e.g. WTIME / HW_ERR / CARD_REMOVAL) + * and is moving to reset state. Any in-flight hardware TX is no longer of + * interest to the FSM; if the driver-level DMA happens to complete later, + * the resulting TX_COMPLETE notification will see tx_busy already false + * and rx_after_tx_compl already false, so it won't spuriously re-enable + * the receiver. */ +void card_uart_tx_abort(struct card_uart *cuart) +{ + OSMO_ASSERT(cuart); + card_uart_wtime_stop(cuart); + cuart->tx_busy = false; + cuart->rx_after_tx_compl = false; +} + int card_uart_tx(struct card_uart *cuart, const uint8_t *data, size_t len, bool rx_after_complete) { OSMO_ASSERT(cuart); diff --git a/ccid_common/cuart.h b/ccid_common/cuart.h index a9afc1d..a53dc18 100644 --- a/ccid_common/cuart.h +++ b/ccid_common/cuart.h @@ -153,6 +153,12 @@ /*! Schedule (asynchronous) transmit data via UART; optionally enable Rx after completion */ int card_uart_tx(struct card_uart *cuart, const uint8_t *data, size_t len, bool rx_after_complete); +/*! Abort any in-flight TX. Clears tx_busy + stops WT timer. Used by the + * ISO7816-3 FSM when transitioning to RESET, so subsequent transactions + * don't trip card_uart_tx's "TX already in flight" assertion on the stale + * tx_busy left behind by an aborted transfer. */ +void card_uart_tx_abort(struct card_uart *cuart); + /*! Schedule (asynchronous) receive data via UART (after CUART_E_RX_COMPLETE) */ int card_uart_rx(struct card_uart *cuart, uint8_t *data, size_t len); diff --git a/ccid_common/iso7816_fsm.c b/ccid_common/iso7816_fsm.c index bad827a..2b3b41e 100644 --- a/ccid_common/iso7816_fsm.c +++ b/ccid_common/iso7816_fsm.c @@ -31,6 +31,7 @@ #include "logging.h" #include "cuart.h" +#include "iso7816_3.h" #include "iso7816_fsm.h" /* unionize to ensure at least properly aligned msgb struct */ @@ -312,7 +313,20 @@ struct iso7816_3_priv *ip = get_iso7816_3_priv(fi); OSMO_ASSERT(fi->fsm == &iso7816_3_fsm); + /* Bring the cuart back to its default per-transaction state, matching + * what card_uart_ctrl(CUART_CTL_POWER_*, 0) does, but without + * power-cycling (warm reset, HW_ERR, WTIME paths reach S_RESET without + * touching power). Any reset path could land here mid-transaction with + * stale state: tx_busy still set from an aborted TX, rx_threshold left + * at e.g. 256 from a multi-byte RX setup, wtime_etu still at a + * PPS-negotiated value rather than the ATR default. Leaving those + * stale breaks the next ATR (next card_uart_tx asserts; or ATR bytes + * pile up in the ringbuffer waiting for a threshold that ATR's max 33 + * bytes can never reach). */ card_uart_ctrl(ip->uart, CUART_CTL_RX_TIMER_HINT, 0); + card_uart_tx_abort(ip->uart); + card_uart_set_rx_threshold(ip->uart, 1); + card_uart_ctrl(ip->uart, CUART_CTL_WTIME, ISO7816_3_DEFAULT_WT); /* go back to initial state in child FSMs */ osmo_fsm_inst_state_chg(ip->atr_fi, ATR_S_WAIT_TS, 0, 0); -- To view, visit https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42779?usp=email To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings?usp=email Gerrit-MessageType: merged Gerrit-Project: osmo-ccid-firmware Gerrit-Branch: master Gerrit-Change-Id: Iac8bd7f4f0eecccc9acce149277a4f5016fec7c1 Gerrit-Change-Number: 42779 Gerrit-PatchSet: 2 Gerrit-Owner: Hoernchen <ewild(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: jolly <andreas(a)eversberg.eu> Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: lynxis lazus <lynxis(a)fe80.eu>

1 0

[S] Change in osmo-ccid-firmware[master]: ccid: reject invalid bClockCommand
by laforge 12 Jun '26

12 Jun '26

laforge has submitted this change. ( https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42778?usp=email ) ( 1 is the latest approved patch-set. No files were changed between the latest approved patch-set and the submitted one. )Change subject: ccid: reject invalid bClockCommand ...................................................................... ccid: reject invalid bClockCommand Change-Id: I34dedcaaaf3cd67e22b207016f08e745736dd625 --- M ccid_common/ccid_device.c 1 file changed, 9 insertions(+), 1 deletion(-) Approvals: Jenkins Builder: Verified jolly: Looks good to me, but someone else must approve laforge: Looks good to me, approved diff --git a/ccid_common/ccid_device.c b/ccid_common/ccid_device.c index 63a4d70..d9faa00 100644 --- a/ccid_common/ccid_device.c +++ b/ccid_common/ccid_device.c @@ -601,9 +601,17 @@ const union ccid_pc_to_rdr *u = msgb_ccid_out(msg); const struct ccid_header *ch = (const struct ccid_header *) u; uint8_t seq = u->icc_clock.hdr.bSeq; + uint8_t cmd = u->icc_clock.bClockCommand; struct msgb *resp; - cs->ci->slot_ops->set_clock(cs, u->icc_clock.bClockCommand); + /* CCID v1.1 §6.1.9: bClockCommand has only two defined values; reject + * early here */ + if (cmd != CCID_CLOCK_CMD_RESTART && cmd != CCID_CLOCK_CMD_STOP) { + resp = ccid_gen_slot_status(cs, seq, CCID_CMD_STATUS_FAILED, offsetof(struct ccid_pc_to_rdr_icc_clock, bClockCommand)); + return ccid_slot_send_unbusy(cs, resp); + } + + cs->ci->slot_ops->set_clock(cs, cmd); resp = ccid_gen_slot_status(cs, seq, CCID_CMD_STATUS_OK, 0); return ccid_slot_send_unbusy(cs, resp); } -- To view, visit https://gerrit.osmocom.org/c/osmo-ccid-firmware/+/42778?usp=email To unsubscribe, or for help writing mail filters, visit https://gerrit.osmocom.org/settings?usp=email Gerrit-MessageType: merged Gerrit-Project: osmo-ccid-firmware Gerrit-Branch: master Gerrit-Change-Id: I34dedcaaaf3cd67e22b207016f08e745736dd625 Gerrit-Change-Number: 42778 Gerrit-PatchSet: 2 Gerrit-Owner: Hoernchen <ewild(a)sysmocom.de> Gerrit-Reviewer: Jenkins Builder Gerrit-Reviewer: jolly <andreas(a)eversberg.eu> Gerrit-Reviewer: laforge <laforge(a)osmocom.org> Gerrit-Reviewer: lynxis lazus <lynxis(a)fe80.eu>

1 0

2026

2025

2024

2023

2022

gerrit-log