{"dataType": "CVE_RECORD", "dataVersion": "5.1", "cveMetadata": {"cveId": "CVE-2024-41003", "assignerOrgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "state": "PUBLISHED", "assignerShortName": "Linux", "dateReserved": "2024-07-12T12:17:45.609Z", "datePublished": "2024-07-12T12:44:39.110Z", "dateUpdated": "2024-11-05T09:34:51.438Z"}, "containers": {"cna": {"providerMetadata": {"orgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "shortName": "Linux", "dateUpdated": "2024-11-05T09:34:51.438Z"}, "descriptions": [{"lang": "en", "value": "In the Linux kernel, the following vulnerability has been resolved:\n\nbpf: Fix reg_set_min_max corruption of fake_reg\n\nJuan reported that after doing some changes to buzzer [0] and implementing\na new fuzzing strategy guided by coverage, they noticed the following in\none of the probes:\n\n [...]\n 13: (79) r6 = *(u64 *)(r0 +0) ; R0=map_value(ks=4,vs=8) R6_w=scalar()\n 14: (b7) r0 = 0 ; R0_w=0\n 15: (b4) w0 = -1 ; R0_w=0xffffffff\n 16: (74) w0 >>= 1 ; R0_w=0x7fffffff\n 17: (5c) w6 &= w0 ; R0_w=0x7fffffff R6_w=scalar(smin=smin32=0,smax=umax=umax32=0x7fffffff,var_off=(0x0; 0x7fffffff))\n 18: (44) w6 |= 2 ; R6_w=scalar(smin=umin=smin32=umin32=2,smax=umax=umax32=0x7fffffff,var_off=(0x2; 0x7ffffffd))\n 19: (56) if w6 != 0x7ffffffd goto pc+1\n REG INVARIANTS VIOLATION (true_reg2): range bounds violation u64=[0x7fffffff, 0x7ffffffd] s64=[0x7fffffff, 0x7ffffffd] u32=[0x7fffffff, 0x7ffffffd] s32=[0x7fffffff, 0x7ffffffd] var_off=(0x7fffffff, 0x0)\n REG INVARIANTS VIOLATION (false_reg1): range bounds violation u64=[0x7fffffff, 0x7ffffffd] s64=[0x7fffffff, 0x7ffffffd] u32=[0x7fffffff, 0x7ffffffd] s32=[0x7fffffff, 0x7ffffffd] var_off=(0x7fffffff, 0x0)\n REG INVARIANTS VIOLATION (false_reg2): const tnum out of sync with range bounds u64=[0x0, 0xffffffffffffffff] s64=[0x8000000000000000, 0x7fffffffffffffff] u32=[0x0, 0xffffffff] s32=[0x80000000, 0x7fffffff] var_off=(0x7fffffff, 0x0)\n 19: R6_w=0x7fffffff\n 20: (95) exit\n\n from 19 to 21: R0=0x7fffffff R6=scalar(smin=umin=smin32=umin32=2,smax=umax=smax32=umax32=0x7ffffffe,var_off=(0x2; 0x7ffffffd)) R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n 21: R0=0x7fffffff R6=scalar(smin=umin=smin32=umin32=2,smax=umax=smax32=umax32=0x7ffffffe,var_off=(0x2; 0x7ffffffd)) R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n 21: (14) w6 -= 2147483632 ; R6_w=scalar(smin=umin=umin32=2,smax=umax=0xffffffff,smin32=0x80000012,smax32=14,var_off=(0x2; 0xfffffffd))\n 22: (76) if w6 s>= 0xe goto pc+1 ; R6_w=scalar(smin=umin=umin32=2,smax=umax=0xffffffff,smin32=0x80000012,smax32=13,var_off=(0x2; 0xfffffffd))\n 23: (95) exit\n\n from 22 to 24: R0=0x7fffffff R6_w=14 R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n 24: R0=0x7fffffff R6_w=14 R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n 24: (14) w6 -= 14 ; R6_w=0\n [...]\n\nWhat can be seen here is a register invariant violation on line 19. After\nthe binary-or in line 18, the verifier knows that bit 2 is set but knows\nnothing about the rest of the content which was loaded from a map value,\nmeaning, range is [2,0x7fffffff] with var_off=(0x2; 0x7ffffffd). When in\nline 19 the verifier analyzes the branch, it splits the register states\nin reg_set_min_max() into the registers of the true branch (true_reg1,\ntrue_reg2) and the registers of the false branch (false_reg1, false_reg2).\n\nSince the test is w6 != 0x7ffffffd, the src_reg is a known constant.\nInternally, the verifier creates a \"fake\" register initialized as scalar\nto the value of 0x7ffffffd, and then passes it onto reg_set_min_max(). Now,\nfor line 19, it is mathematically impossible to take the false branch of\nthis program, yet the verifier analyzes it. It is impossible because the\nsecond bit of r6 will be set due to the prior or operation and the\nconstant in the condition has that bit unset (hex(fd) == binary(1111 1101).\n\nWhen the verifier first analyzes the false / fall-through branch, it will\ncompute an intersection between the var_off of r6 and of the constant. This\nis because the verifier creates a \"fake\" register initialized to the value\nof the constant. The intersection result later refines both registers in\nregs_refine_cond_op():\n\n [...]\n t = tnum_intersect(tnum_subreg(reg1->var_off), tnum_subreg(reg2->var_off));\n reg1->var_o\n---truncated---"}], "affected": [{"product": "Linux", "vendor": "Linux", "defaultStatus": "unaffected", "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git", "programFiles": ["include/linux/bpf_verifier.h", "kernel/bpf/verifier.c"], "versions": [{"version": "67420501e868", "lessThan": "41e8ab428a99", "status": "affected", "versionType": "git"}, {"version": "67420501e868", "lessThan": "92424801261d", "status": "affected", "versionType": "git"}]}, {"product": "Linux", "vendor": "Linux", "defaultStatus": "affected", "repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git", "programFiles": ["include/linux/bpf_verifier.h", "kernel/bpf/verifier.c"], "versions": [{"version": "6.8", "status": "affected"}, {"version": "0", "lessThan": "6.8", "status": "unaffected", "versionType": "semver"}, {"version": "6.9.7", "lessThanOrEqual": "6.9.*", "status": "unaffected", "versionType": "semver"}, {"version": "6.10", "lessThanOrEqual": "*", "status": "unaffected", "versionType": "original_commit_for_fix"}]}], "references": [{"url": "https://git.kernel.org/stable/c/41e8ab428a9964df378fa45760a660208712145b"}, {"url": "https://git.kernel.org/stable/c/92424801261d1564a0bb759da3cf3ccd69fdf5a2"}], "title": "bpf: Fix reg_set_min_max corruption of fake_reg", "x_generator": {"engine": "bippy-9e1c9544281a"}}, "adp": [{"providerMetadata": {"orgId": "af854a3a-2127-422b-91ae-364da2661108", "shortName": "CVE", "dateUpdated": "2024-08-02T04:39:56.068Z"}, "title": "CVE Program Container", "references": [{"url": "https://git.kernel.org/stable/c/41e8ab428a9964df378fa45760a660208712145b", "tags": ["x_transferred"]}, {"url": "https://git.kernel.org/stable/c/92424801261d1564a0bb759da3cf3ccd69fdf5a2", "tags": ["x_transferred"]}]}, {"metrics": [{"other": {"type": "ssvc", "content": {"id": "CVE-2024-41003", "role": "CISA Coordinator", "options": [{"Exploitation": "none"}, {"Automatable": "no"}, {"Technical Impact": "partial"}], "version": "2.0.3", "timestamp": "2024-09-10T17:01:08.608294Z"}}}], "title": "CISA ADP Vulnrichment", "providerMetadata": {"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0", "shortName": "CISA-ADP", "dateUpdated": "2024-09-11T17:34:18.907Z"}}]}}

{"dataType": "CVE_RECORD", "containers": {"adp": [{"title": "CVE Program Container", "references": [{"url": "https://git.kernel.org/stable/c/41e8ab428a9964df378fa45760a660208712145b", "tags": ["x_transferred"]}, {"url": "https://git.kernel.org/stable/c/92424801261d1564a0bb759da3cf3ccd69fdf5a2", "tags": ["x_transferred"]}], "providerMetadata": {"orgId": "af854a3a-2127-422b-91ae-364da2661108", "shortName": "CVE", "dateUpdated": "2024-08-02T04:39:56.068Z"}}, {"title": "CISA ADP Vulnrichment", "metrics": [{"other": {"type": "ssvc", "content": {"id": "CVE-2024-41003", "role": "CISA Coordinator", "options": [{"Exploitation": "none"}, {"Automatable": "no"}, {"Technical Impact": "partial"}], "version": "2.0.3", "timestamp": "2024-09-10T17:01:08.608294Z"}}}], "providerMetadata": {"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0", "shortName": "CISA-ADP", "dateUpdated": "2024-09-11T12:42:21.947Z"}}], "cna": {"title": "bpf: Fix reg_set_min_max corruption of fake_reg", "affected": [{"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git", "vendor": "Linux", "product": "Linux", "versions": [{"status": "affected", "version": "67420501e868", "lessThan": "41e8ab428a99", "versionType": "git"}, {"status": "affected", "version": "67420501e868", "lessThan": "92424801261d", "versionType": "git"}], "programFiles": ["include/linux/bpf_verifier.h", "kernel/bpf/verifier.c"], "defaultStatus": "unaffected"}, {"repo": "https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git", "vendor": "Linux", "product": "Linux", "versions": [{"status": "affected", "version": "6.8"}, {"status": "unaffected", "version": "0", "lessThan": "6.8", "versionType": "custom"}, {"status": "unaffected", "version": "6.9.7", "versionType": "custom", "lessThanOrEqual": "6.9.*"}, {"status": "unaffected", "version": "6.10", "versionType": "original_commit_for_fix", "lessThanOrEqual": "*"}], "programFiles": ["include/linux/bpf_verifier.h", "kernel/bpf/verifier.c"], "defaultStatus": "affected"}], "references": [{"url": "https://git.kernel.org/stable/c/41e8ab428a9964df378fa45760a660208712145b"}, {"url": "https://git.kernel.org/stable/c/92424801261d1564a0bb759da3cf3ccd69fdf5a2"}], "x_generator": {"engine": "bippy-c9c4e1df01b2"}, "descriptions": [{"lang": "en", "value": "In the Linux kernel, the following vulnerability has been resolved:\n\nbpf: Fix reg_set_min_max corruption of fake_reg\n\nJuan reported that after doing some changes to buzzer [0] and implementing\na new fuzzing strategy guided by coverage, they noticed the following in\none of the probes:\n\n [...]\n 13: (79) r6 = *(u64 *)(r0 +0) ; R0=map_value(ks=4,vs=8) R6_w=scalar()\n 14: (b7) r0 = 0 ; R0_w=0\n 15: (b4) w0 = -1 ; R0_w=0xffffffff\n 16: (74) w0 >>= 1 ; R0_w=0x7fffffff\n 17: (5c) w6 &= w0 ; R0_w=0x7fffffff R6_w=scalar(smin=smin32=0,smax=umax=umax32=0x7fffffff,var_off=(0x0; 0x7fffffff))\n 18: (44) w6 |= 2 ; R6_w=scalar(smin=umin=smin32=umin32=2,smax=umax=umax32=0x7fffffff,var_off=(0x2; 0x7ffffffd))\n 19: (56) if w6 != 0x7ffffffd goto pc+1\n REG INVARIANTS VIOLATION (true_reg2): range bounds violation u64=[0x7fffffff, 0x7ffffffd] s64=[0x7fffffff, 0x7ffffffd] u32=[0x7fffffff, 0x7ffffffd] s32=[0x7fffffff, 0x7ffffffd] var_off=(0x7fffffff, 0x0)\n REG INVARIANTS VIOLATION (false_reg1): range bounds violation u64=[0x7fffffff, 0x7ffffffd] s64=[0x7fffffff, 0x7ffffffd] u32=[0x7fffffff, 0x7ffffffd] s32=[0x7fffffff, 0x7ffffffd] var_off=(0x7fffffff, 0x0)\n REG INVARIANTS VIOLATION (false_reg2): const tnum out of sync with range bounds u64=[0x0, 0xffffffffffffffff] s64=[0x8000000000000000, 0x7fffffffffffffff] u32=[0x0, 0xffffffff] s32=[0x80000000, 0x7fffffff] var_off=(0x7fffffff, 0x0)\n 19: R6_w=0x7fffffff\n 20: (95) exit\n\n from 19 to 21: R0=0x7fffffff R6=scalar(smin=umin=smin32=umin32=2,smax=umax=smax32=umax32=0x7ffffffe,var_off=(0x2; 0x7ffffffd)) R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n 21: R0=0x7fffffff R6=scalar(smin=umin=smin32=umin32=2,smax=umax=smax32=umax32=0x7ffffffe,var_off=(0x2; 0x7ffffffd)) R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n 21: (14) w6 -= 2147483632 ; R6_w=scalar(smin=umin=umin32=2,smax=umax=0xffffffff,smin32=0x80000012,smax32=14,var_off=(0x2; 0xfffffffd))\n 22: (76) if w6 s>= 0xe goto pc+1 ; R6_w=scalar(smin=umin=umin32=2,smax=umax=0xffffffff,smin32=0x80000012,smax32=13,var_off=(0x2; 0xfffffffd))\n 23: (95) exit\n\n from 22 to 24: R0=0x7fffffff R6_w=14 R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n 24: R0=0x7fffffff R6_w=14 R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n 24: (14) w6 -= 14 ; R6_w=0\n [...]\n\nWhat can be seen here is a register invariant violation on line 19. After\nthe binary-or in line 18, the verifier knows that bit 2 is set but knows\nnothing about the rest of the content which was loaded from a map value,\nmeaning, range is [2,0x7fffffff] with var_off=(0x2; 0x7ffffffd). When in\nline 19 the verifier analyzes the branch, it splits the register states\nin reg_set_min_max() into the registers of the true branch (true_reg1,\ntrue_reg2) and the registers of the false branch (false_reg1, false_reg2).\n\nSince the test is w6 != 0x7ffffffd, the src_reg is a known constant.\nInternally, the verifier creates a \"fake\" register initialized as scalar\nto the value of 0x7ffffffd, and then passes it onto reg_set_min_max(). Now,\nfor line 19, it is mathematically impossible to take the false branch of\nthis program, yet the verifier analyzes it. It is impossible because the\nsecond bit of r6 will be set due to the prior or operation and the\nconstant in the condition has that bit unset (hex(fd) == binary(1111 1101).\n\nWhen the verifier first analyzes the false / fall-through branch, it will\ncompute an intersection between the var_off of r6 and of the constant. This\nis because the verifier creates a \"fake\" register initialized to the value\nof the constant. The intersection result later refines both registers in\nregs_refine_cond_op():\n\n [...]\n t = tnum_intersect(tnum_subreg(reg1->var_off), tnum_subreg(reg2->var_off));\n reg1->var_o\n---truncated---"}], "providerMetadata": {"orgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "shortName": "Linux", "dateUpdated": "2024-07-15T06:53:15.103Z"}}}, "cveMetadata": {"cveId": "CVE-2024-41003", "state": "PUBLISHED", "dateUpdated": "2024-09-11T17:34:18.907Z", "dateReserved": "2024-07-12T12:17:45.609Z", "assignerOrgId": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "datePublished": "2024-07-12T12:44:39.110Z", "assignerShortName": "Linux"}, "dataVersion": "5.1"}

{"cveTags": [], "descriptions": [{"lang": "en", "value": "In the Linux kernel, the following vulnerability has been resolved:\n\nbpf: Fix reg_set_min_max corruption of fake_reg\n\nJuan reported that after doing some changes to buzzer [0] and implementing\na new fuzzing strategy guided by coverage, they noticed the following in\none of the probes:\n\n [...]\n 13: (79) r6 = *(u64 *)(r0 +0) ; R0=map_value(ks=4,vs=8) R6_w=scalar()\n 14: (b7) r0 = 0 ; R0_w=0\n 15: (b4) w0 = -1 ; R0_w=0xffffffff\n 16: (74) w0 >>= 1 ; R0_w=0x7fffffff\n 17: (5c) w6 &= w0 ; R0_w=0x7fffffff R6_w=scalar(smin=smin32=0,smax=umax=umax32=0x7fffffff,var_off=(0x0; 0x7fffffff))\n 18: (44) w6 |= 2 ; R6_w=scalar(smin=umin=smin32=umin32=2,smax=umax=umax32=0x7fffffff,var_off=(0x2; 0x7ffffffd))\n 19: (56) if w6 != 0x7ffffffd goto pc+1\n REG INVARIANTS VIOLATION (true_reg2): range bounds violation u64=[0x7fffffff, 0x7ffffffd] s64=[0x7fffffff, 0x7ffffffd] u32=[0x7fffffff, 0x7ffffffd] s32=[0x7fffffff, 0x7ffffffd] var_off=(0x7fffffff, 0x0)\n REG INVARIANTS VIOLATION (false_reg1): range bounds violation u64=[0x7fffffff, 0x7ffffffd] s64=[0x7fffffff, 0x7ffffffd] u32=[0x7fffffff, 0x7ffffffd] s32=[0x7fffffff, 0x7ffffffd] var_off=(0x7fffffff, 0x0)\n REG INVARIANTS VIOLATION (false_reg2): const tnum out of sync with range bounds u64=[0x0, 0xffffffffffffffff] s64=[0x8000000000000000, 0x7fffffffffffffff] u32=[0x0, 0xffffffff] s32=[0x80000000, 0x7fffffff] var_off=(0x7fffffff, 0x0)\n 19: R6_w=0x7fffffff\n 20: (95) exit\n\n from 19 to 21: R0=0x7fffffff R6=scalar(smin=umin=smin32=umin32=2,smax=umax=smax32=umax32=0x7ffffffe,var_off=(0x2; 0x7ffffffd)) R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n 21: R0=0x7fffffff R6=scalar(smin=umin=smin32=umin32=2,smax=umax=smax32=umax32=0x7ffffffe,var_off=(0x2; 0x7ffffffd)) R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n 21: (14) w6 -= 2147483632 ; R6_w=scalar(smin=umin=umin32=2,smax=umax=0xffffffff,smin32=0x80000012,smax32=14,var_off=(0x2; 0xfffffffd))\n 22: (76) if w6 s>= 0xe goto pc+1 ; R6_w=scalar(smin=umin=umin32=2,smax=umax=0xffffffff,smin32=0x80000012,smax32=13,var_off=(0x2; 0xfffffffd))\n 23: (95) exit\n\n from 22 to 24: R0=0x7fffffff R6_w=14 R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n 24: R0=0x7fffffff R6_w=14 R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n 24: (14) w6 -= 14 ; R6_w=0\n [...]\n\nWhat can be seen here is a register invariant violation on line 19. After\nthe binary-or in line 18, the verifier knows that bit 2 is set but knows\nnothing about the rest of the content which was loaded from a map value,\nmeaning, range is [2,0x7fffffff] with var_off=(0x2; 0x7ffffffd). When in\nline 19 the verifier analyzes the branch, it splits the register states\nin reg_set_min_max() into the registers of the true branch (true_reg1,\ntrue_reg2) and the registers of the false branch (false_reg1, false_reg2).\n\nSince the test is w6 != 0x7ffffffd, the src_reg is a known constant.\nInternally, the verifier creates a \"fake\" register initialized as scalar\nto the value of 0x7ffffffd, and then passes it onto reg_set_min_max(). Now,\nfor line 19, it is mathematically impossible to take the false branch of\nthis program, yet the verifier analyzes it. It is impossible because the\nsecond bit of r6 will be set due to the prior or operation and the\nconstant in the condition has that bit unset (hex(fd) == binary(1111 1101).\n\nWhen the verifier first analyzes the false / fall-through branch, it will\ncompute an intersection between the var_off of r6 and of the constant. This\nis because the verifier creates a \"fake\" register initialized to the value\nof the constant. The intersection result later refines both registers in\nregs_refine_cond_op():\n\n [...]\n t = tnum_intersect(tnum_subreg(reg1->var_off), tnum_subreg(reg2->var_off));\n reg1->var_o\n---truncated---"}, {"lang": "es", "value": "En el kernel de Linux se ha resuelto la siguiente vulnerabilidad: bpf: Fix reg_set_min_max corrupci\u00f3n de fake_reg Juan inform\u00f3 que despu\u00e9s de hacer algunos cambios al buzzer [0] e implementar una nueva estrategia de fuzzing guiada por cobertura, notaron lo siguiente en una de las sondas : [...] 13: (79) r6 = *(u64 *)(r0 +0) ; R0=map_value(ks=4,vs=8) R6_w=escalar() 14: (b7) r0 = 0 ; R0_w=0 15: (b4) w0 = -1 ; R0_w=0xffffffff 16: (74) w0 >>= 1 ; R0_w=0x7ffffffff 17: (5c) w6 &= w0 ; R0_w=0x7fffffff R6_w=escalar(smin=smin32=0,smax=umax=umax32=0x7fffffff,var_off=(0x0; 0x7fffffff)) 18: (44) w6 |= 2 ; R6_w=scalar(smin=umin=smin32=umin32=2,smax=umax=umax32=0x7fffffff,var_off=(0x2; 0x7ffffffd)) 19: (56) if w6 != 0x7ffffffd goto pc+1 VIOLACI\u00d3N DE INVARIANTES REG (true_reg2) : violaci\u00f3n de l\u00edmites de rango u64=[0x7fffffff, 0x7ffffffd] s64=[0x7fffffff, 0x7ffffffd] u32=[0x7fffffff, 0x7ffffffd] s32=[0x7fffffff, 0x7ffffffd] var_off=(0x7fffffff, 0x0) INVARIANTES DEL REG (false_reg1): violaci\u00f3n de los l\u00edmites de rango u64 =[0x7fffffff, 0x7ffffffd] s64=[0x7fffffff, 0x7ffffffd] u32=[0x7fffffff, 0x7ffffffd] s32=[0x7fffffff, 0x7ffffffd] var_off=(0x7fffffff, 0x0) VIOLACI\u00d3N DE INVARIANTES REG (false_reg2): st tnum no est\u00e1 sincronizado con los l\u00edmites de rango u64 =[0x0, 0xffffffffffffffff] s64=[0x80000000000000000, 0x7ffffffffffffff] u32=[0x0, 0xffffffff] s32=[0x80000000, 0x7ffffff] var_off=(0x7ffffffff, 0x0) 19: 6_w=0x7fffffff 20: (95) salida del 19 al 21: R0=0x7fffffff R6=escalar(smin=umin=smin32=umin32=2,smax=umax=smax32=umax32=0x7ffffffe,var_off=(0x2; 0x7ffffffd)) R7=map_ptr(ks=4,vs=8) R9=ctx () R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm 21: R0=0x7fffffff R6=escalar(smin=umin=smin32=umin32=2,smax=umax=smax32=umax32 =0x7ffffffe,var_off=(0x2; 0x7ffffffd)) R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm 21: (14) w6 -= 2147483632; R6_w=scalar(smin=umin=umin32=2,smax=umax=0xffffffff,smin32=0x80000012,smax32=14,var_off=(0x2; 0xfffffffd)) 22: (76) si w6 s>= 0xe goto pc+1; R6_w=scalar(smin=umin=umin32=2,smax=umax=0xffffffff,smin32=0x80000012,smax32=13,var_off=(0x2; 0xfffffffd)) 23: (95) salida de 22 a 24: R0=0x7fffffff R6_w= 14 R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm 24: R0=0x7ffffffff R6_w=14 R7= map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm 24: (14) w6 -= 14 ; R6_w=0 [...] Lo que se puede ver aqu\u00ed es una violaci\u00f3n de invariante de registro en la l\u00ednea 19. Despu\u00e9s del binario o en la l\u00ednea 18, el verificador sabe que el bit 2 est\u00e1 establecido pero no sabe nada sobre el resto del contenido que fue cargado desde un valor de mapa, es decir, el rango es [2,0x7fffffff] con var_off=(0x2; 0x7ffffffd). Cuando en la l\u00ednea 19 el verificador analiza la rama, divide los estados de registro en reg_set_min_max() en los registros de la rama verdadera (true_reg1, true_reg2) y los registros de la rama falsa (false_reg1, false_reg2). Dado que la prueba es w6! = 0x7ffffffd, src_reg es una constante conocida. Internamente, el verificador crea un registro \"falso\" inicializado como escalar al valor de 0x7ffffffd y luego lo pasa a reg_set_min_max(). Ahora, para la l\u00ednea 19, es matem\u00e1ticamente imposible tomar la rama falsa de este programa, sin embargo, el verificador la analiza. Es imposible porque el segundo bit de r6 se establecer\u00e1 debido a la operaci\u00f3n anterior o y la constante en la condici\u00f3n tiene ese bit sin configurar (hex(fd) == binario(1111 1101). Cuando el verificador analiza por primera vez el valor falso/ca\u00edda -a trav\u00e9s de la rama, calcular\u00e1 una intersecci\u00f3n entre el var_off de r6 y la constante. Esto se debe a que el verificador crea un registro \"falso\" inicializado con el valor de la constante. El resultado de la intersecci\u00f3n luego refina ambos registros en regs_refine_cond_op(): [...] t = tnum_intersect(tnum_subreg(reg1->var_off), tnum_subreg(reg2->var_off));"}], "id": "CVE-2024-41003", "lastModified": "2024-07-12T16:34:58.687", "metrics": {}, "published": "2024-07-12T13:15:21.180", "references": [{"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "url": "https://git.kernel.org/stable/c/41e8ab428a9964df378fa45760a660208712145b"}, {"source": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "url": "https://git.kernel.org/stable/c/92424801261d1564a0bb759da3cf3ccd69fdf5a2"}], "sourceIdentifier": "416baaa9-dc9f-4396-8d5f-8c081fb06d67", "vulnStatus": "Awaiting Analysis"}

{"bugzilla": {"description": "kernel: bpf: Fix reg_set_min_max corruption of fake_reg", "id": "2297587", "url": "https://bugzilla.redhat.com/show_bug.cgi?id=2297587"}, "csaw": false, "cvss3": {"cvss3_base_score": "4.4", "cvss3_scoring_vector": "CVSS:3.1/AV:L/AC:L/PR:H/UI:N/S:U/C:N/I:N/A:H", "status": "draft"}, "cwe": "CWE-99", "details": ["In the Linux kernel, the following vulnerability has been resolved:\nbpf: Fix reg_set_min_max corruption of fake_reg\nJuan reported that after doing some changes to buzzer [0] and implementing\na new fuzzing strategy guided by coverage, they noticed the following in\none of the probes:\n[...]\n13: (79) r6 = *(u64 *)(r0 +0) ; R0=map_value(ks=4,vs=8) R6_w=scalar()\n14: (b7) r0 = 0 ; R0_w=0\n15: (b4) w0 = -1 ; R0_w=0xffffffff\n16: (74) w0 >>= 1 ; R0_w=0x7fffffff\n17: (5c) w6 &= w0 ; R0_w=0x7fffffff R6_w=scalar(smin=smin32=0,smax=umax=umax32=0x7fffffff,var_off=(0x0; 0x7fffffff))\n18: (44) w6 |= 2 ; R6_w=scalar(smin=umin=smin32=umin32=2,smax=umax=umax32=0x7fffffff,var_off=(0x2; 0x7ffffffd))\n19: (56) if w6 != 0x7ffffffd goto pc+1\nREG INVARIANTS VIOLATION (true_reg2): range bounds violation u64=[0x7fffffff, 0x7ffffffd] s64=[0x7fffffff, 0x7ffffffd] u32=[0x7fffffff, 0x7ffffffd] s32=[0x7fffffff, 0x7ffffffd] var_off=(0x7fffffff, 0x0)\nREG INVARIANTS VIOLATION (false_reg1): range bounds violation u64=[0x7fffffff, 0x7ffffffd] s64=[0x7fffffff, 0x7ffffffd] u32=[0x7fffffff, 0x7ffffffd] s32=[0x7fffffff, 0x7ffffffd] var_off=(0x7fffffff, 0x0)\nREG INVARIANTS VIOLATION (false_reg2): const tnum out of sync with range bounds u64=[0x0, 0xffffffffffffffff] s64=[0x8000000000000000, 0x7fffffffffffffff] u32=[0x0, 0xffffffff] s32=[0x80000000, 0x7fffffff] var_off=(0x7fffffff, 0x0)\n19: R6_w=0x7fffffff\n20: (95) exit\nfrom 19 to 21: R0=0x7fffffff R6=scalar(smin=umin=smin32=umin32=2,smax=umax=smax32=umax32=0x7ffffffe,var_off=(0x2; 0x7ffffffd)) R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n21: R0=0x7fffffff R6=scalar(smin=umin=smin32=umin32=2,smax=umax=smax32=umax32=0x7ffffffe,var_off=(0x2; 0x7ffffffd)) R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n21: (14) w6 -= 2147483632 ; R6_w=scalar(smin=umin=umin32=2,smax=umax=0xffffffff,smin32=0x80000012,smax32=14,var_off=(0x2; 0xfffffffd))\n22: (76) if w6 s>= 0xe goto pc+1 ; R6_w=scalar(smin=umin=umin32=2,smax=umax=0xffffffff,smin32=0x80000012,smax32=13,var_off=(0x2; 0xfffffffd))\n23: (95) exit\nfrom 22 to 24: R0=0x7fffffff R6_w=14 R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n24: R0=0x7fffffff R6_w=14 R7=map_ptr(ks=4,vs=8) R9=ctx() R10=fp0 fp-24=map_ptr(ks=4,vs=8) fp-40=mmmmmmmm\n24: (14) w6 -= 14 ; R6_w=0\n[...]\nWhat can be seen here is a register invariant violation on line 19. After\nthe binary-or in line 18, the verifier knows that bit 2 is set but knows\nnothing about the rest of the content which was loaded from a map value,\nmeaning, range is [2,0x7fffffff] with var_off=(0x2; 0x7ffffffd). When in\nline 19 the verifier analyzes the branch, it splits the register states\nin reg_set_min_max() into the registers of the true branch (true_reg1,\ntrue_reg2) and the registers of the false branch (false_reg1, false_reg2).\nSince the test is w6 != 0x7ffffffd, the src_reg is a known constant.\nInternally, the verifier creates a \"fake\" register initialized as scalar\nto the value of 0x7ffffffd, and then passes it onto reg_set_min_max(). Now,\nfor line 19, it is mathematically impossible to take the false branch of\nthis program, yet the verifier analyzes it. It is impossible because the\nsecond bit of r6 will be set due to the prior or operation and the\nconstant in the condition has that bit unset (hex(fd) == binary(1111 1101).\nWhen the verifier first analyzes the false / fall-through branch, it will\ncompute an intersection between the var_off of r6 and of the constant. This\nis because the verifier creates a \"fake\" register initialized to the value\nof the constant. The intersection result later refines both registers in\nregs_refine_cond_op():\n[...]\nt = tnum_intersect(tnum_subreg(reg1->var_off), tnum_subreg(reg2->var_off));\nreg1->var_o\n---truncated---"], "name": "CVE-2024-41003", "package_state": [{"cpe": "cpe:/o:redhat:enterprise_linux:6", "fix_state": "Not affected", "package_name": "kernel", "product_name": "Red Hat Enterprise Linux 6"}, {"cpe": "cpe:/o:redhat:enterprise_linux:7", "fix_state": "Out of support scope", "package_name": "kernel", "product_name": "Red Hat Enterprise Linux 7"}, {"cpe": "cpe:/o:redhat:enterprise_linux:7", "fix_state": "Out of support scope", "package_name": "kernel-rt", "product_name": "Red Hat Enterprise Linux 7"}, {"cpe": "cpe:/o:redhat:enterprise_linux:8", "fix_state": "Not affected", "package_name": "kernel", "product_name": "Red Hat Enterprise Linux 8"}, {"cpe": "cpe:/o:redhat:enterprise_linux:8", "fix_state": "Not affected", "package_name": "kernel-rt", "product_name": "Red Hat Enterprise Linux 8"}, {"cpe": "cpe:/o:redhat:enterprise_linux:9", "fix_state": "Will not fix", "package_name": "kernel", "product_name": "Red Hat Enterprise Linux 9"}, {"cpe": "cpe:/o:redhat:enterprise_linux:9", "fix_state": "Will not fix", "package_name": "kernel-rt", "product_name": "Red Hat Enterprise Linux 9"}], "public_date": "2024-07-12T00:00:00Z", "references": ["https://www.cve.org/CVERecord?id=CVE-2024-41003\nhttps://nvd.nist.gov/vuln/detail/CVE-2024-41003\nhttps://lore.kernel.org/linux-cve-announce/2024071244-CVE-2024-41003-792f@gregkh/T"], "threat_severity": "Low", "upstream_fix": "kernel 6.9.7, kernel 6.10-rc5"}