From acc2d9330719ed8b4829693e659df696e214d59e Mon Sep 17 00:00:00 2001
From: vasilito <adminpupkin@gmail.com>
Date: Tue, 30 Jun 2026 14:51:06 +0300
Subject: [PATCH] fix: PCI summary JSON parser to match real QEMU info pci
 output

The awk patterns in hw_gen_pci_json() targeted a hypothetical output format
that never matched QEMU's actual HMP 'info pci' response:

- /^Bus / anchored to line start, but QEMU outputs '  Bus  ' (2 leading spaces)
- /vendor_id = / expected 'vendor_id = 0x8086', but QEMU prints 'PCI device 8086:2922'
- /device_id = / expected separate line, but vendor:device are on the same line
- /class = / expected 'class = 0x010601', but QEMU prints 'SATA controller:' or 'Class 0106:'
- IRQ field expected 'IRQ 0.' but QEMU prints 'IRQ 5, pin A'

Result: pci_summary.json was always invalid JSON with orphaned key-value pairs.

Fix verified against QEMU source (pci-hmp-cmds.c:31-51) and tested with
realistic 4-device output: all devices correctly parsed with bus/dev/fn,
vendor_id, device_id, and IRQ fields.

Found by: 5-agent parallel review (QA execution agent)
---
 host/initramfs/hw_collect.sh | 36 ++++++++++++++++++++++--------------
 1 file changed, 22 insertions(+), 14 deletions(-)

diff --git a/host/initramfs/hw_collect.sh b/host/initramfs/hw_collect.sh
index 7bd6580..d2689ca 100755
--- a/host/initramfs/hw_collect.sh
+++ b/host/initramfs/hw_collect.sh
@@ -444,29 +444,37 @@ hw_gen_pci_json() {
         printf '  "devices": [\n'
 
         awk '
-            /^Bus / {
+            # Device header: "  Bus  0, device   1, function 0:"
+            # (QEMU outputs 2 leading spaces — no ^ anchor)
+            /Bus .*device .*function/ {
                 if (found) printf " },\n"
                 found = 1
                 bus = $2; gsub(/,/, "", bus)
                 dev = $4; gsub(/,/, "", dev)
                 fn = $6; gsub(/:/, "", fn)
                 printf "    {\"bus\": %s, \"device\": %s, \"function\": %s", bus, dev, fn
+                next
             }
-            /vendor_id = / {
-                gsub(/.*= /, "", $0)
-                printf ", \"vendor_id\": \"%s\"", $NF
+            # Vendor/device line: "    SATA controller: PCI device 8086:2922"
+            # or:                   "    Class 0106: PCI device 8086:2922"
+            /PCI device/ {
+                for (i = 1; i <= NF; i++) {
+                    if (index($i, ":") > 0) {
+                        split($i, vd, ":")
+                        if (length(vd[1]) == 4 && length(vd[2]) == 4 && \
+                            vd[1] ~ /^[0-9a-fA-F]+$/ && vd[2] ~ /^[0-9a-fA-F]+$/) {
+                            printf ", \"vendor_id\": \"0x%s\", \"device_id\": \"0x%s\"", vd[1], vd[2]
+                            break
+                        }
+                    }
+                }
+                next
             }
-            /device_id = / {
-                gsub(/.*= /, "", $0)
-                printf ", \"device_id\": \"%s\"", $NF
-            }
-            /class = / {
-                gsub(/.*= /, "", $0)
-                printf ", \"class\": \"%s\"", $NF
-            }
-            /IRQ / {
-                _irq = $2; gsub(/\./, "", _irq)
+            # IRQ line: "      IRQ 5, pin A"
+            /IRQ [0-9]/ {
+                _irq = $2; gsub(/,/, "", _irq)
                 printf ", \"irq\": \"%s\"", _irq
+                next
             }
             END { if (found) printf " }"; printf "\n" }
         ' "$_pci" 2>/dev/null